An integrated mortise and tenon structure building intelligent design method based on three-dimensional modeling

Abstract

The application relates to the technical field of intelligent design, and discloses an integrated mortise and tenon structure building intelligent design method based on three-dimensional modeling, which comprises the following steps: extracting the slope exposure coefficient and the direction opening degree coefficient of a candidate node to convert an equivalent water content fluctuation amplitude; setting the tenon length, the mortise depth and the cross section ratio in a three-dimensional initial model as design variables to calculate an embedded margin; combining a bamboo gradient amplification coefficient to calculate a directional embedded attenuation amount; using the attenuation amount to reduce the initial stiffness of the node to obtain an equivalent lateral spring stiffness; after standardizing the design variables, establishing a search space with the overall lateral stiffness as a target; using an optimization algorithm to iteratively standardize the parameters and dynamically update the initial model, calculating the overall lateral stiffness until convergence to obtain optimal parameters; and restoring the optimal parameters to actual processing dimensions.

Description

Technical Field

[0001] This invention relates to the field of intelligent design technology, and more specifically, to an intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling. Background Technology

[0002] With the popularization of green building concepts, integrated mortise and tenon building systems based on 3D parametric modeling and CNC machining have received widespread attention. In the current automated design and optimization process, optimization algorithms such as particle swarm optimization are usually used to improve the nominal geometric stiffness of the connection parts as the sole guide, iteratively optimizing spatial parameters such as tenon length and mortise depth, and then directly outputting CNC machining instructions.

[0003] However, existing conventional intelligent design methods reveal significant inherent flaws when faced with real and complex outdoor service environments:

[0004] First, both wood and bamboo exhibit extremely high humidity sensitivity. In real construction sites, different facades experience significant directional humidity fluctuations due to variations in terrain slope, windwardness, and openness. Existing 3D modeling and optimization frameworks often assume an absolutely ideal, uniform external environment, completely ignoring the continuous expansion of gaps induced by humidity fluctuations along specific directions. Second, when mortise and tenon joints are applied to bamboo components, the vascular bundles within the bamboo exhibit a gradient property: a dense, hard outer layer and a relatively sparse inner layer. Existing design systems typically treat materials as isotropic or homogeneous entities, failing to reflect the amplifying effect of this material gradient distribution on gap loosening. Because the directional attenuation induced by the environment and the natural material differences of bamboo are not incorporated into the closed-loop optimization process, the optimal parameters derived by existing methods are often merely ideal initial forms in digital simulation space. When the system is actually put into extreme external environments, its embedment margin will be rapidly consumed over the service life, causing the actual lateral stiffness of the frame to drop sharply when facing extreme lateral loads, which can easily lead to overall instability and deformation. Summary of the Invention

[0005] This invention provides an intelligent design method for integrated mortise and tenon structure buildings based on 3D modeling, which solves the technical problems mentioned in the background art.

[0006] This invention provides an intelligent design method for integrated mortise and tenon structure buildings based on 3D modeling, comprising:

[0007] Extract the slope exposure coefficient and directional openness coefficient of the candidate mortise and tenon joints, and convert the slope exposure coefficient and directional openness coefficient into equivalent moisture content fluctuation amplitude;

[0008] A three-dimensional initial model is established, and the tenon length, mortise depth and cross-sectional ratio of the candidate mortise and tenon joints in the three-dimensional initial model are set as design variables to calculate the mortise and tenon fitment margin.

[0009] Extract the equivalent elastic modulus and vascular bundle distribution ratio of the components connected to the candidate mortise and tenon joints to establish the bamboo gradient amplification factor; combine the bamboo gradient amplification factor, the equivalent moisture content fluctuation amplitude and the mortise and tenon embedment margin to calculate the directional embedment attenuation.

[0010] Obtain the initial stiffness of the candidate mortise and tenon joint; reduce the initial stiffness using the directional embedment attenuation to obtain the equivalent lateral spring stiffness;

[0011] The design variables are standardized to obtain standardized design parameters, and a search space is established with the overall lateral stiffness as the target.

[0012] The standardized design parameters are iterated using an optimization algorithm to dynamically update the three-dimensional initial model, and the equivalent lateral spring stiffness is assembled. The overall lateral stiffness is calculated until convergence to obtain the optimal parameters.

[0013] The optimal parameters are restored to the actual machining dimensions, and a tool center trajectory corresponding to the actual machining dimensions is generated and output based on the surface normal of the candidate mortise and tenon nodes.

[0014] The beneficial effects of this invention are as follows: By extracting the parameters of site exposure and openness to calculate the equivalent moisture content fluctuation, and combining the bamboo gradient amplification coefficient of connected components with the tenon and mortise embedding margin, a directional embedding attenuation amount is constructed. Then, the initial stiffness of the node is scientifically reduced to the equivalent lateral spring stiffness. And by using an optimization algorithm to perform a comprehensive iteration in the search space of the overall lateral stiffness of the assembly, the optimal parameters and the corresponding surface normal machining tool trajectory are directly output. This invention breaks through the limitation of traditional design that only pursues the maximization of nominal geometric stiffness. It incorporates the humidity-induced gap attenuation and the internal material difference of bamboo in the real service environment into the closed-loop optimization, realizing the deep integration of digital space optimization and maintaining the maximization of stiffness under real service conditions. At the same time, it breaks down the barrier from three-dimensional parametric generation to automated CNC machining, significantly improving the extreme lateral load resistance and intelligent construction efficiency of the integrated tenon and mortise building system. Attached Figure Description

[0015] Figure 1 This is a flowchart of an intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling, according to the present invention.

[0016] Figure 2 This is a schematic diagram of a specific implementation scenario of the present invention. Detailed Implementation

[0017] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0018] like Figure 1 As shown, an intelligent design method for integrated mortise and tenon structure buildings based on 3D modeling includes:

[0019] Extract the slope exposure coefficient and directional openness coefficient of the candidate mortise and tenon joints, and convert the slope exposure coefficient and directional openness coefficient into equivalent moisture content fluctuation amplitude;

[0020] A three-dimensional initial model is established, and the tenon length, mortise depth and cross-sectional ratio of the candidate mortise and tenon joints in the three-dimensional initial model are set as design variables to calculate the mortise and tenon fitment margin.

[0021] Extract the equivalent elastic modulus and vascular bundle distribution ratio of the components connected to the candidate mortise and tenon joints to establish the bamboo gradient amplification factor; combine the bamboo gradient amplification factor, the equivalent moisture content fluctuation amplitude and the mortise and tenon embedment margin to calculate the directional embedment attenuation.

[0022] Obtain the initial stiffness of the candidate mortise and tenon joint; reduce the initial stiffness using the directional embedment attenuation to obtain the equivalent lateral spring stiffness;

[0023] The design variables are standardized to obtain standardized design parameters, and a search space is established with the overall lateral stiffness as the target.

[0024] The standardized design parameters are iterated using an optimization algorithm to dynamically update the three-dimensional initial model, and the equivalent lateral spring stiffness is assembled. The overall lateral stiffness is calculated until convergence to obtain the optimal parameters.

[0025] The optimal parameters are restored to the actual machining dimensions, and a tool center trajectory corresponding to the actual machining dimensions is generated and output based on the surface normal of the candidate mortise and tenon nodes.

[0026] Preferably, the slope exposure coefficient and directional openness coefficient of the candidate mortise and tenon joints are extracted, and the slope exposure coefficient and directional openness coefficient are converted into equivalent moisture content fluctuation amplitude, including:

[0027] The slope exposure coefficient, the directional openness coefficient, and the equivalent moisture content fluctuation amplitude are calculated using the following formulas:

[0028]

[0029]

[0030]

[0031] In the formula, The slope exposure coefficient is mentioned above. The vertical component of the average normal corresponding to the candidate mortise and tenon joint; The directional openness coefficient is the value of the direction. This represents the number of unit vectors normal to the local terrain. For quantity serial number, Let be the unit vector normal to the local terrain. The unit vector of the facade where the candidate mortise and tenon joint is located faces the direction of the extreme lateral load; The equivalent moisture content fluctuation amplitude, The amplitude of the site's baseline moisture content fluctuation. This is the openness weighting coefficient. This is the slope weighting coefficient. For windward weighting coefficient, This represents the exposure coefficient in extreme directions.

[0032] The local terrain normal unit vector is a dimensionless vector that characterizes the spatial orientation of the local terrain surface, reflecting the actual tilt direction of the terrain. It can be extracted using neighborhood principal component analysis of the site's 3D point cloud data. First, outlier removal and coordinate registration are performed on the site's point cloud. Then, the point cloud neighborhood corresponding to the node is selected, the neighborhood covariance matrix is ​​calculated, and the eigenvectors are solved to obtain the corresponding terrain normal unit vector.

[0033] The unit vector of the facade containing the candidate mortise and tenon joint facing the direction of the extreme lateral load is a dimensionless vector representing the building facade to which the candidate mortise and tenon joint belongs facing the direction of the extreme lateral load flow, reflecting the relative spatial orientation of the load and the building facade. The principal direction of the load can be determined through extreme load analysis of the building structure, and then the unit vector of the facade pointing in the direction of the load can be calculated by combining the spatial orientation coordinates of the building facade.

[0034] The vertical component of the average normal corresponding to the candidate mortise and tenon node is the dimensionless numerical component of the average normal unit vector of the terrain in the corresponding area of ​​the candidate mortise and tenon node, reflecting the vertical inclination of the terrain. The average normal unit vector of the terrain can be obtained by extracting multiple terrain normal unit vectors from the area corresponding to the node and calculating their average value, and then extracting the numerical component of this vector on the vertical coordinate axis.

[0035] The number of local terrain normal unit vectors is the number of local terrain normal unit vectors selected for calculating the directional openness coefficient. It is a positive integer, reflecting the terrain sample size involved in the openness calculation. It can be determined by the density of the site's ground cloud and the size of the projection area of ​​the corresponding facade of the node. The terrain normal unit vectors covering this projection area are selected to ensure the engineering representativeness of the sample size.

[0036] The site reference moisture content fluctuation amplitude is a dimensionless, decimal-based value characterizing the basic fluctuation difference in moisture content of timber and bamboo substrates in a building site from their processed state to their extreme service state. It serves as the benchmark value for calculating nodal moisture content fluctuations. It can be obtained by collecting long-term climate monitoring data from the site, combining the design value of the processed moisture content of timber and bamboo with the measured equilibrium moisture content of the substrate under extreme climate conditions, and calculating the difference between the two.

[0037] The extreme direction exposure coefficient is a dimensionless coefficient characterizing the windward exposure of the facade where a candidate mortise and tenon joint is located under extreme lateral loads, reflecting the load distribution ratio on each joint facade. It can be obtained by performing linear elastic calculations of the building structure under extreme working conditions, reading the load distribution values ​​of the extreme lateral loads on each joint facade, and then comparing each distribution value with the maximum distribution value.

[0038] The openness weighting coefficient is a dimensionless coefficient used to quantify the influence of the directional openness coefficient on the amplitude of equivalent moisture content fluctuation. It is used to balance the role of the openness factor in the calculation of moisture content fluctuation. A value of 0.30 is preferred, as this value reasonably reflects the actual impact of site openness on the nodal moisture content fluctuation without excessively amplifying the amplitude of moisture content fluctuation due to the openness factor.

[0039] The slope weighting coefficient is a dimensionless coefficient used to quantify the influence of the slope exposure coefficient on the amplitude of equivalent moisture content fluctuation, and is used to balance the role of slope factors in moisture content fluctuation calculation. A value of 0.15 is preferred, as this value matches the actual influence of terrain slope on node humidity exposure and avoids deviations in moisture content fluctuation calculation caused by slope factors.

[0040] The windward weighting coefficient is a dimensionless coefficient used to quantify the influence of the extreme direction exposure coefficient on the amplitude of equivalent moisture content fluctuation. It is used to balance the role of the windward effect in the calculation of moisture content fluctuation. A value of 0.25 is preferred, as this value reasonably reflects the amplification effect of the load on the humidity exposure of the nodes in the windward direction, consistent with the humidity response law of bamboo and wood substrates in actual engineering.

[0041] The slope exposure coefficient is a dimensionless coefficient that characterizes the influence of the terrain slope of the candidate mortise and tenon joint on the moisture exposure of the joint. It reflects the degree of amplification of moisture content fluctuation caused by terrain slope and is calculated from the vertical component of the average normal corresponding to the candidate mortise and tenon joint.

[0042] The directional openness coefficient is a dimensionless coefficient that characterizes the influence of the spatial openness of the facade where the candidate mortise and tenon joint is located on the humidity exposure of the joint. It reflects the amplification of moisture content fluctuation caused by the openness of the site and is calculated from the operation results of the local topographic normal unit vector and the load direction unit vector.

[0043] The equivalent moisture content fluctuation amplitude is a dimensionless parameter representing the actual moisture content fluctuation amplitude of timber and bamboo faced by candidate mortise and tenon joints under the combined effects of site topographic slope, spatial openness, and extreme load windward effect. It is the core parameter for calculating the subsequent directional gap growth of the joint and is obtained by multiplying the site reference moisture content fluctuation amplitude by the combination coefficient.

[0044] In detail, the specific implementation first involves calculating the dot product of the local terrain normal unit vector and the load direction unit vector. After eliminating invalid values ​​with negative dot product results, the summation and average are calculated to obtain the directional openness coefficient, which reflects the actual openness of the site. Then, by subtracting the absolute value of the average vertical component of the terrain normal from 1, the slope exposure coefficient, reflecting the terrain slope exposure, is obtained. Subsequently, the two coefficients and the extreme directional exposure coefficient are multiplied by their corresponding weighting coefficients, and then summed with 1 to obtain the combination coefficient. Finally, the actual equivalent moisture content fluctuation amplitude at the node is obtained by multiplying the site's benchmark moisture content fluctuation amplitude by the combination coefficient. This method accurately converts geometric and load characteristics such as terrain slope openness, load windward effect, etc. The humidity fluctuation parameters can be directly used in structural design. For example, if the average normal vertical component of the terrain in the area corresponding to a building node is 0.6, the slope exposure coefficient can be calculated to be 0.4. If 100 local terrain normal unit vectors are selected in the area corresponding to the node, and the sum of the maximum values ​​after the dot product with the load direction unit vector is 50, the directional openness coefficient can be calculated to be 0.5. Combined with the openness weight coefficient of 0.30, the slope weight coefficient of 0.15, the windward weight coefficient of 0.25, and the extreme direction exposure coefficient of 0.8, the combination coefficient can be calculated to be 1.43. If the site reference moisture content fluctuation amplitude is 0.04, the equivalent moisture content fluctuation amplitude of the node can be obtained as 0.0572.

[0045] In detail, when extracting the normal unit vector of the local terrain, firstly, statistical outlier removal is performed on the 3D point cloud of the site. After removing outliers exceeding 3 times the standard deviation, coordinate registration and voxel downsampling are performed. Then, adaptive neighborhood analysis is used, with the neighborhood radius being the maximum of the average point distance of the point cloud after 3 times downsampling and 0.02 meters. If the number of neighborhood points is less than 30, the radius is gradually increased to 30 to 80. Then, the normal unit vector is extracted through principal component analysis. The unit vector of the facade where the candidate mortise and tenon joint is located facing the extreme lateral load direction is first determined through the lateral resistance design of the building structure. Then, based on the planar coordinates and spatial orientation of the building facade, the unit vector pointing from the facade to the load principal direction is calculated. The fluctuation amplitude of the site's reference moisture content is selected as the difference between the moisture content in the processing state and the moisture content in the design extreme service state, expressed in decimal form. For example, if the processing moisture content is 12% and the extreme service moisture content is 12%, then the difference is 12%. If the site's benchmark moisture content is 8%, then the fluctuation amplitude of the site's benchmark moisture content is taken as 0.04. The extreme direction exposure coefficient is obtained by calculating the linear elasticity of the building structure under extreme working conditions. After reading the load distribution value of the facade where each node is located, the distribution value of a single node is compared with the maximum distribution value of all nodes. The openness, slope, and windward weight coefficients are initially calculated using the recommended initial values ​​of 0.30, 0.15, and 0.25. If the obtained equivalent moisture content fluctuation amplitude exceeds the reasonable range of 0.01 to 0.1, it is uniformly multiplied by the normalization coefficient to adjust it to this range. Subsequent secondary calibration can be performed based on the project's on-site monitoring data. The number of local terrain normal unit vectors should cover the projection area of ​​the facade corresponding to the node, with a minimum number of no less than 50, to ensure the stability of the calculation results. For example, if the projection area of ​​the facade corresponding to the node is 5 square meters and the point cloud density is 20 points per square meter, then 100 local terrain normal unit vectors are selected for calculation.

[0046] Preferably, a three-dimensional initial model is established, and the tenon length, mortise depth, and cross-sectional ratio of the candidate mortise and tenon joints in the three-dimensional initial model are set as design variables to calculate the mortise and tenon fitment margin, including:

[0047] The equivalent width, equivalent height, and tenon-tenon fitment margin of the candidate mortise and tenon joint are calculated according to the following formulas:

[0048]

[0049]

[0050]

[0051] In the formula, The equivalent width is... For the target cross-sectional area, The cross-sectional ratio is mentioned. The equivalent height; The tenon and mortise fitting allowance, The length of the tenon, The depth of the mortise is given.

[0052] The target cross-sectional area is the effective cross-sectional area of ​​the candidate mortise and tenon joint as preset in the structural design. It can be obtained by calculating the envelope value of the axial compression and shear load requirements along the grain of the building structure, and determined in combination with the compressive and shear strength of bamboo and wood materials.

[0053] The cross-sectional ratio is the ratio of the equivalent width to the equivalent height of a candidate mortise and tenon joint, and it is dimensionless. It can be determined by the stress direction of the building structure and the spatial arrangement requirements of the joints, and selected in conjunction with the feasibility of CNC machining.

[0054] The tenon length is the effective extension length of the tenon in a mortise and tenon joint. It can be determined by the joint fastening requirements of the building structure and the cross-sectional dimensions of the components, combined with the mechanical properties of bamboo and wood materials.

[0055] The mortise depth is the effective embedment depth of the mortise in a mortise and tenon joint. It can be determined by the tenon length and the required embedment reserve of the joint, and matched with the tool travel limitations of CNC machining.

[0056] The equivalent width is the effective width within the force-bearing plane of the node, calculated from the target cross-sectional area and cross-sectional ratio, reflecting the force-bearing size of the node in the width direction.

[0057] The equivalent height is the effective height within the force-bearing plane of the node, calculated from the target cross-sectional area and cross-sectional ratio, reflecting the force-bearing dimensions of the node in the height direction.

[0058] The tenon and mortise fitment margin is a node fitment reserve characteristic quantity calculated from the tenon length, mortise depth, equivalent height and cross-sectional ratio, reflecting the geometric fitment capacity of the node to resist gap expansion.

[0059] In detail, the specific implementation involves first determining the target cross-sectional area of ​​the node based on the axial compression and shear load requirements along the grain of the building structure. Then, considering the force direction and processing requirements of the node, a cross-sectional ratio is selected. The equivalent width is obtained by multiplying the target cross-sectional area by the cross-sectional ratio and taking the square root. The equivalent height is obtained by dividing the target cross-sectional area by the cross-sectional ratio and taking the square root. Next, the first calculated value is obtained by multiplying the tenon length by the mortise depth and dividing the result by the equivalent height. The second calculated value is obtained by multiplying the cross-sectional ratio by 2 and dividing the result by 1 and adding it to the cross-sectional ratio. Finally, the mortise and tenon joint is obtained by multiplying the two calculated values. The method of mortise and tenon joint retention margin transforms abstract geometric parameters into structural retention features, avoiding the distortion of retention capacity caused by single parameter optimization. For example, if the target cross-sectional area of ​​a candidate mortise and tenon joint is 0.0144 square meters and the cross-sectional ratio is 0.8, the calculated equivalent width is 0.1073 meters and the equivalent height is 0.1342 meters. The tenon length of the joint is 0.09 meters and the mortise depth is 0.10 meters. The calculated first value is 0.0671 meters and the second value is 0.8889. Multiplying the two values ​​gives a mortise and tenon retention margin of 0.0596 meters.

[0060] In detail, the target cross-sectional area is first calculated based on the envelope value of two types of load requirements: axial compression and shear along the grain. The axial compression requirement is calculated by dividing the design value of the axial compression at the node by the product of the material's compressive strength and the utilization factor. The shear requirement is calculated by multiplying the design value of the shear force along the grain at the node by the shear shape correction factor, and then dividing by the product of the material's shear strength along the grain and the utilization factor. The utilization factor is preferably set to 0.7, and the shear shape correction factor is preferably set to 1.1. The maximum value of these two calculation results is taken as the initial target cross-sectional area. If subsequent structural simulations show that the cross-sectional dimensions are unreasonable, a secondary correction can be performed, and this value should be kept constant throughout an optimization batch. The tenon length is limited to 0.6 to 1.4 times the initial tenon length, and the mortise depth is not less than and not more than 1.25 times the tenon length to avoid ineffective fastening due to excessively short tenons and shallow mortises. The reasonable range for the cross-sectional ratio is limited to 0.5 to 1.8 to avoid excessively large or small cross-sectional ratios that result in overly flat or thin node cross-sections, affecting the stress and processing performance of the node. When the equivalent height is in a single extreme lateral load direction, it is taken as the cross-sectional dimension perpendicular to that load direction. When there are two orthogonal extreme lateral load directions, the equivalent height and equivalent width are used as height parameters for different directions to calculate the tenon and mortise fitment margin. The more unfavorable value in the calculation results is taken as the fitment margin of the node. The reasonable engineering range for the tenon and mortise fitment margin is 0.02 to 0.2 meters. If the calculation result exceeds this range, it can be corrected to a reasonable range by adjusting the tenon length or mortise depth.

[0061] Preferably, extracting the equivalent elastic modulus and vascular bundle distribution ratio of the components connected to the candidate mortise and tenon joints to establish the bamboo gradient amplification factor includes: calculating the bamboo gradient amplification factor according to the following formula:

[0062]

[0063] In the formula, The gradient amplification factor of the bamboo material; The equivalent elastic modulus of the outer layer. The equivalent elastic modulus of the inner layer; The proportion of outer vascular bundles. The vascular bundle distribution ratio is the ratio of the inner layer vascular bundles; the equivalent elastic modulus of the outer layer and the equivalent elastic modulus of the inner layer together constitute the equivalent elastic modulus, and the vascular bundle distribution ratio of the outer layer vascular bundles and the vascular bundle distribution ratio of the inner layer vascular bundles together constitute the vascular bundle distribution ratio.

[0064] The outer layer equivalent elastic modulus is the equivalent elastic modulus of the outer layer region of the bamboo component connected by the candidate mortise and tenon joint, reflecting the stiffness characteristics of the outer layer of the bamboo component. It can be obtained by conducting mechanical tensile tests on specimens from the outer layer region of the bamboo component, or by retrieving corresponding values ​​from a material property database of the same bamboo species.

[0065] The equivalent elastic modulus of the inner layer is the equivalent elastic modulus of the inner layer region of a bamboo component connected by a candidate mortise and tenon joint, reflecting the stiffness characteristics of the inner layer of the bamboo component. It can be obtained through mechanical tensile testing of specimens from the inner layer region of the bamboo component, or by retrieving corresponding values ​​from a material property database of the same bamboo species.

[0066] The distribution ratio of outer vascular bundles refers to the volume or area fraction of vascular bundles in the outer region of bamboo components connected by candidate mortise and tenon joints, reflecting the density of vascular bundles in the outer layer of the bamboo component. This ratio can be obtained through statistical analysis of microscopic scanning images of the bamboo component's cross-section, or manually calculated using cross-sectional dissection.

[0067] The distribution ratio of vascular bundles in the inner layer refers to the volume or area fraction of vascular bundles in the inner region of bamboo components connected by candidate mortise and tenon joints, reflecting the density of vascular bundles in the inner layer of the bamboo component. This ratio can be obtained through statistical analysis of microscopic scanning images of the bamboo component's cross-section, or manually calculated using cross-sectional dissection.

[0068] The bamboo gradient amplification factor is a dimensionless factor obtained by integrating the equivalent elastic modulus of the outer and inner layers of bamboo components and the vascular bundle distribution ratio. It reflects the degree to which the gradient properties of bamboo amplify the attenuation of node embedment.

[0069] In detail, the bamboo component wall thickness is first divided radially, and the equivalent elastic modulus and vascular bundle distribution ratio of the outer and inner layers are extracted. Then, the equivalent elastic modulus of the outer layer is divided by the equivalent elastic modulus of the inner layer, and the square root of the result is taken to obtain the first characteristic value. This process ensures that the modulus gradient has a strong influence on the amplification factor. Next, the vascular bundle distribution ratio of the outer layer is divided by the vascular bundle distribution ratio of the inner layer, and the fourth root of the result is taken to obtain the second characteristic value. This process allows the vascular bundle ratio gradient to gently amplify the amplification factor, avoiding the material gradient value from overshadowing the previous value. The effect of the variable is determined by multiplying the two characteristic values ​​to obtain the bamboo gradient amplification coefficient, thus realizing the cross-scale transformation from the microscopic material characteristics of bamboo to the macroscopic structural design factors. For example, if the equivalent elastic modulus of the outer layer of a bamboo component is 12,000 MPa and the equivalent elastic modulus of the inner layer is 6,000 MPa, the first characteristic value is calculated to be approximately 1.414. The vascular bundle distribution ratio of the outer layer of the bamboo component is 0.6 and the vascular bundle distribution ratio of the inner layer is 0.2, resulting in a second characteristic value of approximately 1.316. Multiplying the two values ​​together yields the bamboo gradient amplification coefficient of approximately 1.861.

[0070] In detail, the radial division standard for the outer and inner layers of bamboo components is to divide the wall thickness of the bamboo component radially into four equal layers, taking the outer quarter layer as the outer layer region and the inner quarter layer as the inner layer region. This division method conforms to the radial distribution law of bamboo vascular bundle density. When measuring the equivalent elastic modulus of the outer and inner layers, the number of samples for each group of bamboo components should not be less than 5, and the average value of the measured values ​​of the samples is taken as the final parameter value. If there is no actual measurement condition, typical values ​​from the database of bamboo species in the same region can be used. When statistically analyzing the distribution ratio of vascular bundles in the outer and inner layers, volume fraction is preferred. If there is no volume fraction statistical condition, the area fraction of the cross section can be used as an approximation. When conducting the statistics, three different cross sections at the connection between the bamboo component and the node are selected, and the average value of the statistical results is taken. When the node is a pure wood component without bamboo gradient properties, the bamboo gradient amplification factor is directly set to 1. In this case, the factor no longer amplifies the node embedment attenuation. All bamboo parameters are collected from the effective connection area between the bamboo component and the mortise and tenon node to avoid parameters from non-stressed areas affecting the accuracy of the calculation results.

[0071] Preferably, the directional fixation attenuation is calculated by combining the bamboo gradient amplification factor, the equivalent moisture content fluctuation amplitude, and the tenon and mortise fixation margin, including:

[0072] The directional gap growth and the directional embedment attenuation are calculated using the following formulas:

[0073]

[0074]

[0075] In the formula, The directional gap growth amount, This is the radial shrinkage coefficient of wood. For equivalent height, The amplitude of the equivalent moisture content fluctuation; This refers to the directional embedding attenuation. The gradient amplification factor of the bamboo material is given. The tenon and mortise fitting allowance.

[0076] The radial shrinkage coefficient of wood is a dimensionless coefficient characterizing the radial shrinkage deformation of wood when its moisture content changes. It reflects the degree of moisture sensitivity of wood. It can be obtained through radial shrinkage tests on the corresponding wood species, or by retrieving standard values ​​for the same type of wood from a wood material performance database.

[0077] The directional gap growth is a value calculated from the radial shrinkage coefficient of the wood, the equivalent height, and the amplitude of the equivalent moisture content fluctuation. It reflects the degree of gap expansion of the mortise and tenon joint along a specific facade direction induced by site humidity fluctuations.

[0078] Directional embedment attenuation is a dimensionless characteristic quantity obtained by integrating the bamboo gradient amplification coefficient, directional gap growth, and mortise and tenon embedment margin. It reflects the degree of consumption of the geometric embedment reserve by the gap growth induced by humidity at the mortise and tenon joint, and the comprehensive result of the degree of consumption after being amplified by the bamboo gradient stiffness.

[0079] In detail, the specific implementation first obtains the corresponding radial shrinkage coefficient of the wood according to the type of wood used in the mortise and tenon joint. Then, this coefficient is multiplied by the equivalent height of the joint and the equivalent moisture content fluctuation amplitude to obtain the humidity-induced directional gap growth. Next, the directional gap growth is divided by the mortise and tenon embedment margin to obtain the basic consumption ratio of gap growth to embedment reserve. Finally, this ratio is multiplied by the bamboo gradient amplification coefficient to obtain the final directional embedment attenuation. This method integrates multiple influencing factors into a single quantitative feature, and the result is dimensionless and can be directly incorporated into subsequent optimization models. For example, if the radial shrinkage coefficient of the wood in a certain mortise and tenon joint is 0.0012, the equivalent height is 0.12 meters, and the equivalent moisture content fluctuation amplitude is 0.04, the calculated directional gap growth is 0.0000576 meters. The mortise and tenon embedment margin of this joint is 0.0596 meters. The bamboo gradient amplification coefficient is 1.861, and the calculated directional embedment attenuation is approximately 0.00179.

[0080] In detail, when selecting the radial shrinkage coefficient of timber, it is necessary to match the actual type of timber used in the mortise and tenon joint and the moisture content range of its application. A shrinkage coefficient value with a moisture content between 8% and 15% should be preferred, as this range aligns with the typical service moisture content range for bamboo and wood construction. Engineering verification of directional gap growth can be completed through humidity aging tests on mortise and tenon joints of the same size. The measured gap growth is compared with the calculated value; if the deviation exceeds 10%, a linear correction is made to the radial shrinkage coefficient of the timber. The reasonable engineering range for directional embedment attenuation is 0 to 0.05. If the calculated result exceeds this range, the embedment margin of the mortise and tenon joint can be increased by adjusting geometric parameters such as tenon length and mortise depth to correct it to a reasonable range. When a joint is simultaneously subjected to humidity fluctuations in multiple directions, the directional embedment attenuation in each direction is calculated separately, and the maximum value is taken as the final embedment attenuation of the joint.

[0081] Preferably, the initial stiffness of the candidate tenon joint is reduced by the directional embedment attenuation to obtain an equivalent lateral spring stiffness, including:

[0082] The initial stiffness, the rotational stiffness, and the equivalent lateral spring stiffness are calculated using the following formulas:

[0083]

[0084]

[0085]

[0086] In the formula, The initial stiffness, For local equivalent compression modulus, For equivalent width, For equivalent height, The length of the tenon; To maintain rotational stiffness, These are the node topology correction coefficients. The attenuation sensitivity coefficient, This refers to the directional embedding attenuation. The depth of the mortise; The equivalent lateral spring stiffness is... The lever arm height.

[0087] The local equivalent compression modulus is the equivalent elastic modulus of bamboo and wood materials at the candidate mortise and tenon joint connection under compression stress, reflecting the stiffness characteristics of the compression contact area of ​​the joint. It can be obtained through mechanical testing of bamboo and wood materials under compression stress, or estimated through the material's compressive strength and shear strength along the grain when no actual testing is available.

[0088] The lever arm height is the vertical distance from the candidate mortise and tenon joint to the corresponding inter-story load-bearing reference plane, reflecting the lever arm effect of the joint rotational stiffness on the overall lateral stiffness of the building. It can be obtained by measuring the spatial coordinates of the building's three-dimensional parametric model and determining the specific value by combining it with the load-bearing story division of the building structure.

[0089] The node topology correction coefficient is a dimensionless coefficient for stiffness correction adapted to different types of mortise and tenon joints, reflecting the actual influence of the structural form of the mortise and tenon joint on its rotational stiffness. The preferred values ​​are 1.00 for straight tenon joints, 1.08 for dovetail joints, 1.03 for through tenon joints, and 0.95 for bamboo-wood hybrid saddle joints. These values ​​accurately match the actual embedment efficiency of various types of joints.

[0090] The attenuation sensitivity coefficient is a dimensionless coefficient characterizing the influence of directional embedment attenuation on the reduction of node stiffness, reflecting the sensitivity of node stiffness to embedment attenuation. A value of 2.03 is preferred. When the directional embedment attenuation is 0.08, the rotational stiffness of the node is reduced to 0.85 times the nominal initial stiffness. This attenuation ratio meets the conventional stiffness attenuation design requirements for bamboo and wood mortise and tenon structures.

[0091] Initial stiffness is the nominal rotational stiffness of the candidate mortise and tenon joint without considering factors such as embedment attenuation and joint type. It reflects the basic stiffness capacity determined by the joint geometry and material properties.

[0092] The characteristic attenuation coefficient is a dimensionless coefficient calculated from the attenuation sensitivity coefficient and the directional embedded attenuation, reflecting the degree of nonlinear reduction of nodal stiffness by directional embedded attenuation.

[0093] The depth reduction factor is a dimensionless coefficient calculated from the mortise depth and tenon length, reflecting the degree to which the length matching relationship between the mortise and tenon corrects the node fastening efficiency.

[0094] Maintaining rotational stiffness is the actual rotational stiffness of a node after considering node type, embedment attenuation, and mortise length matching, reflecting the core embedment stiffness capability of the node under actual service conditions.

[0095] The equivalent lateral spring stiffness is the conversion of the node's rotational stiffness into a lateral stiffness that can be directly used in the simulation of the overall building structure, realizing the dimensional transformation from the local rotational stiffness of the node to the overall lateral stiffness.

[0096] In detail, the specific implementation first obtains parameters such as the local equivalent clamping modulus and the node topology correction coefficient. The local equivalent clamping modulus is multiplied by the equivalent width, then multiplied by the cube of the equivalent height. The result is divided by the product of the tenon length and 12 to obtain the initial node stiffness. Next, the attenuation sensitivity coefficient is multiplied by the directional clamping attenuation and the negative value is taken. An exponential operation is performed with the natural constant as the base to obtain the characteristic attenuation coefficient. The mortise depth is divided by the sum of the mortise depth and the tenon length to obtain the depth reduction coefficient. Subsequently, the node topology correction coefficient, initial stiffness, characteristic attenuation coefficient, and depth reduction coefficient are multiplied sequentially to obtain the node holding rotation stiffness. Finally, the holding rotation stiffness is divided by the square of the lever arm height to obtain the equivalent lateral spring stiffness. This method ensures that the calculated node stiffness closely matches the attenuation characteristics of actual service and is compatible with the overall structural simulation system. For example, if a node has a local equivalent compression modulus of 8000 MPa, an equivalent width of 0.1073 m, an equivalent height of 0.1342 m, and a tenon length of 0.09 m, the calculated initial stiffness is approximately 1473 Nm. With a node topology correction factor of 1.00, an attenuation sensitivity factor of 2.03, a directional attenuation of 0.00179, the calculated characteristic attenuation factor is approximately 0.996. With a mortise depth of 0.10 m, the calculated depth reduction factor is approximately 0.526. Multiplying these together, the calculated rotational stiffness is approximately 780 Nm. With a lever arm height of 2.8 m, the calculated equivalent lateral spring stiffness is approximately 98 Nm.

[0097] In detail, when no actual measurement conditions are available for the local equivalent compression modulus, it is estimated by taking the square root of the product of the material's compressive strength and its shear strength along the grain, and then multiplying it by a dimensionless correction factor of 50 to 120. A correction factor of 85 is preferred. The deviation between this estimated value and the measured value can be controlled within the allowable range for engineering. The node topology correction factor can be calibrated a second time during the project implementation phase through mechanical tests on similar nodes. If the deviation between the measured stiffness and the calculated value exceeds 15%, the factor is linearly adjusted. The attenuation sensitivity factor can be calibrated according to the stiffness attenuation target of the engineering design. The desired stiffness retention ratio is set as the benchmark value, and the specific value of the factor is calculated back through logarithmic operations to meet the stiffness design requirements of different projects. The selection of the lever arm height is determined according to the building structure form. For single-story frames, it is the vertical distance from the node to the roof control point; for multi-story frames, it is the vertical distance from the node to the reference surface of the rigid floor slab of the current floor; for arch-shell hybrid systems, it is the shortest lever arm distance from the node to the equivalent lateral displacement control section, ensuring that the lever arm height matches the actual stress on the structure. When the equivalent lateral spring stiffness is assembled into the overall building stiffness matrix, it is directly incorporated into the diagonal position of the corresponding node degree of freedom. If it is a component connection node, it is incorporated into the connection matrix of the corresponding element.

[0098] Preferably, the design variables are standardized to obtain standardized design parameters, and a search space is established with the overall lateral stiffness as the target, including:

[0099] The standardized design parameters and optimization target values ​​are calculated according to the following formulas, and the boundary constraints of the search space are constructed:

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106] In the formula, Standardized parameters for tenons, The length of the tenon, This is the reference length for the tenon; Standardized parameters for the mortise eye. The depth of the mortise is [not specified]. For reference depth of the mortise; For proportional standardization parameters, The cross-sectional ratio is... The cross-sectional reference ratio is used; the standardized parameters of the tenon, the standardized parameters of the mortise, and the standardized parameters of the ratio together constitute the standardized design parameters. The target value for optimization is... The overall lateral stiffness is... For reference lateral stiffness; This is the maximum mortise depth; For equivalent width, For the tool radius, This is the machining allowance.

[0107] The tenon reference length is a benchmark dimension used for standardizing tenon lengths. It can be obtained by extracting the median of the tenon lengths of all candidate mortise and tenon joints in the initial 3D architectural model, or by selecting the benchmark tenon length value from the initial design scheme.

[0108] The mortise reference depth is a benchmark used for mortise depth standardization. It can be obtained by extracting the median mortise depth of all candidate mortise and tenon joints in the initial 3D building model, or by selecting the mortise depth benchmark value from the initial design scheme.

[0109] The cross-sectional reference scale is a benchmark scale used for cross-sectional scale standardization. It can be obtained by extracting the median of the cross-sectional scales of all candidate mortise and tenon joints in the initial 3D building model, or it can be determined by selecting the benchmark value of the cross-sectional scale from the building structural design.

[0110] The reference lateral stiffness is the overall lateral stiffness of the initial 3D topological model of the building, and serves as the benchmark stiffness for numerical calculations of the optimization objective. It can be calculated by assembling the stiffness matrix of the initial model and then solving for the displacement response using extreme load vectors.

[0111] The tool radius is the effective radius of the tool used for CNC machining of mortise and tenon joints. It can be obtained by consulting the product technical parameters of CNC machining tools, or by measuring the actual dimensions of the tool.

[0112] The maximum mortise depth is the upper limit of the design for mortise depth. It is preferably 1.25 times the length of the corresponding tenon. This is because of the material strength characteristics of bamboo and wood components and the limitations of CNC machining tool stroke, so as to avoid excessive weakening of the strength of the component base material due to excessively deep mortises, while ensuring the engineering rationality of mortise and tenon fastening.

[0113] Machining allowance is the dimensional allowance reserved during CNC machining of mortise and tenon joints to ensure machining accuracy and fit performance. It is preferably 0.002 meters, with a range of 0.001 meters to 0.005 meters, due to the cutting characteristics of bamboo and wood materials and the conventional accuracy requirements of CNC machining, to avoid machining errors leading to mortise and tenon joint failure.

[0114] The standard parameter for tenons is the ratio of the tenon length to the tenon reference length, and it is dimensionless.

[0115] The mortise standardization parameter is the ratio of the mortise depth to the mortise reference depth, and it is dimensionless.

[0116] The standardized proportional parameter is the ratio of the cross-sectional scale to the cross-sectional reference scale, and it is dimensionless.

[0117] Standardized design parameters are a dimensionless set of parameters consisting of tenon standardized parameters, mortise standardized parameters, and proportional standardized parameters.

[0118] The target value for optimization is the ratio of the overall lateral stiffness to the reference lateral stiffness, which is dimensionless.

[0119] Overall lateral stiffness is the actual lateral stiffness of a building structure under extreme working conditions after considering the attenuation of nodal embedment.

[0120] The minimum machining width is the minimum machinable width of the mortise and tenon joint calculated from the tool radius and machining allowance, and it serves as the basis for determining machining constraints in the search space.

[0121] In detail, the specific implementation involves first extracting the reference length of the tenon, the reference depth of the mortise, and the reference scale of the cross-section from the initial 3D building model. Simultaneously, the reference lateral stiffness of the initial model is calculated. The tool radius is obtained by consulting the machining tool parameters, and machining allowances are set. Then, the tenon length, mortise depth, and cross-section scale are divided by their corresponding reference dimensions to obtain three standardized parameters, which are then combined to form standardized design parameters. The ratio of the overall lateral stiffness to the reference lateral stiffness is used as the optimization target value. A search space is constructed with maximizing this value as the guiding principle. Geometric constraints are then set in the search space, limiting the tenon length to no greater than the mortise depth and the mortise depth to no greater than the maximum mortise depth. Finally, the tool radius is multiplied by 2 and then added... The minimum machining width is obtained by determining the allowance. The equivalent width is then limited to this value. This method eliminates the influence of dimensions on the optimization algorithm and ensures that the optimization result combines structural performance and engineering manufacturability. For example, for a node with a tenon length of 0.09 meters, a reference tenon length of 0.08 meters, a standardized tenon parameter of 1.125, a mortise depth of 0.10 meters, a reference mortise depth of 0.09 meters, a standardized mortise parameter of 1.111, a cross-sectional ratio of 0.8, a reference cross-sectional ratio of 0.75, a standardized ratio parameter of 1.067, a tool radius of 0.003 meters, a machining allowance of 0.002 meters, and a minimum machining width of 0.008 meters, limiting the equivalent width to not less than this value satisfies the machining requirements.

[0122] In detail, the reference length of the tenon, the reference depth of the mortise, and the reference scale of the section are all preferably selected from the median of the parameters corresponding to all candidate nodes in the initial 3D building model. This ensures that the reference scale has statistical representativeness and avoids standardization bias caused by a single benchmark value. The reference lateral stiffness is obtained by assembling the linear elastic stiffness of the beam-column bamboo pole units and the initial stiffness of the nodes in the initial model to obtain the overall stiffness matrix. After substituting the load vector under extreme working conditions to solve for the displacement, it is calculated by the ratio of the load to the displacement of the control layer. If the maximum mortise depth cannot be 1.25 times the tenon length due to component size limitations, it can be adjusted to 0.3 to 0.5 times the component section height, while ensuring that it is not less than the tenon length. The machining allowance can be adjusted according to the CNC machining accuracy, taking 0.001 meters to 0.002 meters for finishing and 0.003 meters to 0.005 meters for roughing. In addition to geometric and machining constraints, stress constraints must be included in the search space. The nodal compressive stress must not exceed the material's design compressive strength, and the nodal shear stress along the grain must not exceed the material's design shear strength along the grain. Geometric and machining constraints are applied first, followed by stress constraints. The standardized design parameters are limited to a range of 0.6 to 1.4 to avoid the optimization algorithm finding extreme values ​​without engineering significance. The valid criterion for the optimization target value is greater than 1, meaning the optimized overall lateral stiffness must be higher than the initial model. If the iteration result is less than 1, the design variable boundaries are readjusted.

[0123] Preferably, the standardized design parameters are iteratively evaluated using an optimization algorithm to dynamically update the three-dimensional initial model, and the equivalent lateral spring stiffness is assembled. The overall lateral stiffness is calculated until convergence to obtain the optimal parameters, including:

[0124] The standardized design parameters are iterated according to the following formula to calculate the overall lateral stiffness and convergence error:

[0125]

[0126]

[0127]

[0128]

[0129]

[0130] In the formula, To improve the speed of parameter updates for the next generation, For inertial weights, To ensure the current generation of parameter update speed, For individual learning factors, The first random number, For the individual's historical optimal parameter, These are current-generation standardized design parameters; As a group learning factor, The second random number, The parameter is the historical best for the group; Standardized design parameters for the next generation; For the overall stiffness matrix, It is a displacement vector. This represents the load vector under extreme operating conditions. The overall lateral stiffness is... For the control layer displacement vector; To minimize convergence error, The current optimal target value is set as follows. This is the optimal target value for the previous generation.

[0131] The extreme load vector is a load distribution vector characterizing the various stress-bearing parts of a building structure under extreme lateral loads, reflecting the load state under extreme conditions. It can be obtained through extreme load combination analysis of the lateral resistance design of the building structure, and the specific value can be determined by combining the calculation specifications for extreme loads such as site wind loads and seismic loads.

[0132] Inertia weight is a dimensionless coefficient used in particle swarm optimization (PSO) to balance global search and local convergence capabilities, affecting the degree of decay in parameter update speed. Ideally, it should linearly decay from 0.9 to 0.4, a value that ensures global search range in the early stages of iteration and improves local convergence accuracy in the later stages.

[0133] The individual learning factor is a dimensionless coefficient in particle swarm optimization (PSO) that characterizes the degree to which a single particle learns from its own historical best position, affecting the algorithm's local search capability. A value of 1.8 is preferred, as this ensures the convergence rate of particles towards their optimal solution without causing the algorithm to prematurely fall into local optima.

[0134] The swarm learning factor is a dimensionless coefficient in particle swarm optimization (PSO) that characterizes the degree to which an individual particle learns from the swarm's historical best position, influencing the algorithm's global search capability. An optimal value of 1.8 is preferred, as this value allows particles to quickly approach the global optimum while simultaneously ensuring the algorithm's iterative stability by matching the individual learning factor.

[0135] The first random number is a value used in the particle swarm optimization algorithm to increase the randomness of individual learning items and improve the diversity of the algorithm's search. The preferred value range is 0 to 1. The uncertainty of the random number can prevent the algorithm from iterating on a fixed path and reduce the probability of local optima.

[0136] The second random number is a value used in the particle swarm optimization algorithm to increase the randomness of the swarm learning terms and improve the comprehensiveness of the algorithm's global search. The preferred value range is 0 to 1. The second random number, in conjunction with the first random number, enriches the search paths for algorithm iterations, ensuring comprehensive optimization.

[0137] The current generation parameter update rate is the update rate of the standardized design parameters in the current iteration of the particle swarm optimization algorithm. It is dimensionless and reflects the magnitude of parameter change in the current generation.

[0138] The next-generation parameter update rate is the dimensionless parameter update rate of the next-generation iteration of the particle swarm optimization algorithm, calculated from the inertia weight term, individual learning term, and swarm learning term.

[0139] The current generation standardized design parameters are dimensionless design parameters in the current iteration of the particle swarm optimization algorithm, reflecting the state of the node design parameters in the current generation.

[0140] The next-generation standardized design parameters are dimensionless design parameters obtained by adding the update rates of the current-generation standardized design parameters and the next-generation parameters. They represent the node design parameter states after the particle swarm algorithm iterations.

[0141] The individual historical optimal parameter is the optimal standardized design parameter obtained by a single particle in each iteration of the particle swarm optimization algorithm. It is dimensionless and is the iterative optimization objective of a single particle.

[0142] The swarm historical optimal parameter is the optimal standardized design parameter obtained by all particles in each iteration of the particle swarm optimization algorithm. It is dimensionless and is the global iterative optimization target of the entire algorithm.

[0143] The overall stiffness matrix of generation t is the stiffness matrix obtained by assembling the linear elastic stiffness of the building beam-column bamboo pole elements and the equivalent lateral spring stiffness of the nodes in the t-th iteration of the particle swarm algorithm.

[0144] The displacement vector of generation t is the displacement vector of each part of the building structure obtained by solving the overall stiffness matrix and the extreme load vector in the t-th iteration of the particle swarm optimization algorithm. It reflects the deformation state of the structure under the parameters of that generation.

[0145] The displacement vector of the t-th generation control layer is the displacement vector of the control layer of the building structure extracted from the displacement vector of the t-th generation, reflecting the overall lateral deformation degree of the structure.

[0146] The overall lateral stiffness of generation t is the actual lateral stiffness of the building structure calculated by the extreme load vector and the control layer displacement vector in the t-th iteration of the particle swarm optimization algorithm, reflecting the structural lateral performance under the parameters of that generation.

[0147] The optimal target value for the current generation is the maximum value among all the target values ​​for all particles in the current iteration of the particle swarm optimization algorithm. It is dimensionless and reflects the optimal structural performance of that iteration.

[0148] The previous generation's optimal target value is the maximum value among all the target values ​​corresponding to all particles in the previous iteration of the particle swarm optimization algorithm, and it is dimensionless.

[0149] The convergence error of generation t is a dimensionless relative error calculated from the numerical values ​​of the current generation and the previous generation's optimal search objective.

[0150] The optimal parameter is the population historical optimal parameter obtained after the particle swarm algorithm has converged iteratively. It is dimensionless and is a nodal design parameter that reflects the optimal lateral resistance performance of the building structure.

[0151] In detail, the specific implementation first obtains algorithm parameters such as inertia weights and learning factors, as well as extreme load vectors. The difference between the individual historical best parameters and the current generation standardized design parameters is calculated, and combined with the individual learning factor and the first random number to obtain the first update term. Then, the difference between the group's historical best parameters and the current generation standardized design parameters is calculated, and combined with the group learning factor and the second random number to obtain the second update term. Subsequently, the product of the inertia weights and the current generation parameter update rate is added to the two update terms to obtain the next generation parameter update rate, which is then added to the current generation standardized design parameters to obtain the next generation standardized design parameters. Next, based on the equivalent lateral spring stiffness, the t-th generation overall stiffness matrix is ​​assembled, and substituted into the extreme load vector to solve for the t-th generation displacement vector and extract the control. The overall lateral stiffness of generation t is calculated by the ratio of the absolute values ​​of the load vector and the control layer displacement vector. Then, the convergence error is obtained by calculating the relative error between the current generation and the previous generation's optimal target value. When the convergence error meets the preset condition, the group's historical best parameter is taken as the optimal parameter. This method allows the optimization algorithm to be deeply integrated with structural simulation, avoiding the disconnect between numerical optimization and engineering reality. For example, in a certain generation iteration, the target value corresponding to the group's historical best parameter is 1.35, the current generation's best is 1.352, and the previous generation's best is 1.349. The calculated convergence error is about 0.0022. If the preset convergence condition is 0.003, it is determined that the generation meets the convergence trend. The iteration continues until multiple consecutive generations meet the condition to determine the optimal parameter.

[0152] In detail, the inertia weight is adjusted using a linear decay method, gradually decreasing from 0.9 at the beginning of the iteration to 0.4 at the end of the iteration, with the decay amplitude of each generation evenly distributed according to the maximum iteration number. The first and second random numbers are generated using a uniform random number generation method, randomly selecting values ​​within the range of 0 to 1, and the two random numbers are generated independently without affecting each other. If the extreme working condition load vector needs to consider two orthogonal lateral load directions, the overall lateral stiffness in both directions is calculated separately, and the smaller value is taken as the optimization target value for the particle. The assembly of the overall stiffness matrix integrates the linear elastic stiffness of the beam-column bamboo pole unit and the equivalent lateral spring stiffness of the mortise and tenon joint. The nodal spring stiffness is directly incorporated into the diagonal position of the corresponding degree of freedom, and the component unit stiffness is assembled to the corresponding position using the finite element method. The control layer displacement vector preferentially selects the vertex displacement or bottom displacement of the building structure, and for multi-story buildings, the inter-story displacement of each floor is selected as an auxiliary control basis. The convergence condition adopts a dual judgment standard: if the convergence error is less than 0.001 and there is no better optimization target value for 8 consecutive generations, the iteration is terminated immediately. The particle swarm optimization algorithm uses 30 to 50 particles and a maximum number of iterations of 80 to 150 generations. This parameter range ensures optimization accuracy while controlling computational cost. When multiple parameters are considered equally optimal, the parameter with the smaller building-level embedment attenuation is prioritized, followed by the parameter with the smaller total mortise depth, and finally the parameter with the smaller total tenon length. After particle iteration exceeds the boundary, a projection back to the boundary acceleration attenuation method is used. The standardized parameters that exceed the boundary are projected to the nearest reasonable boundary, and the corresponding velocity components are updated to -0.4 times their original values. The structural displacement response is solved using the stiffness matrix method, obtaining the displacement vector by directly solving the linear equations.

[0153] Preferably, the optimal parameters are restored to the actual machining dimensions, and a tool center trajectory corresponding to the actual machining dimensions is generated and output based on the surface normal of the candidate mortise and tenon nodes, including:

[0154] The actual machining dimensions and the tool center trajectory points are calculated using the following formula:

[0155]

[0156]

[0157]

[0158]

[0159]

[0160]

[0161] In the formula, is the optimal tenon length. Here, represents the reference length for the tenon; represents the optimal standardized parameter for the tenon; and represents the optimal mortise depth. Here, represents the reference depth for the mortise; represents the optimal standardized parameter for the mortise; and represents the optimal cross-sectional ratio. The cross-sectional reference ratio is denoted as , and the optimal ratio standardization parameter is denoted as . The optimal tenon standardization parameter, the optimal mortise standardization parameter, and the optimal ratio standardization parameter together constitute the optimal parameter; is the optimal equivalent width. The target cross-sectional area is denoted as ; the optimal equivalent height is denoted as ; the optimal tenon length, the optimal mortise depth, the optimal cross-sectional ratio, the optimal equivalent width, and the optimal equivalent height together constitute the actual machining dimensions. The tool center trajectory point is used to generate the tool center trajectory; These are the parameter curve points of the surface to be processed; This refers to the arc length parameter of the curve. The radius of the cutting tool; Let be the surface normal unit vector, which is used as the surface normal.

[0162] The optimal tenon standardization parameter is the standardized dimensionless parameter corresponding to the tenon length in the global optimal parameters after the particle swarm algorithm has converged iteratively.

[0163] The optimal normalized parameter for the mortise is the normalized dimensionless parameter corresponding to the mortise depth in the global optimal parameters after the particle swarm optimization algorithm has converged iteratively.

[0164] The optimal proportional standardized parameter is the standardized dimensionless parameter of the corresponding cross-sectional proportion in the global optimal parameters after the particle swarm algorithm has converged iteratively.

[0165] The optimal tenon length is the optimal actual physical length of the tenon joint obtained by multiplying the optimal tenon standardization parameter by the tenon reference length.

[0166] The optimal mortise depth is the optimal actual physical depth of the mortise joint obtained by multiplying the optimal mortise normalization parameter by the mortise reference depth.

[0167] The optimal cross-sectional ratio is the width-to-height ratio of the mortise and tenon joint obtained by multiplying the optimal ratio standardization parameter by the cross-sectional reference ratio.

[0168] The optimal equivalent width is the optimal actual effective width in the force plane of the mortise and tenon joint, calculated from the target cross-sectional area and the optimal cross-sectional ratio.

[0169] The optimal equivalent height is the optimal actual effective height within the force plane of the mortise and tenon joint, calculated from the target cross-sectional area and the optimal cross-sectional ratio.

[0170] The actual machining dimensions are a complete set of machining dimensions for mortise and tenon joints, consisting of the optimal tenon length, optimal mortise depth, optimal cross-sectional ratio, optimal equivalent width, and optimal equivalent height.

[0171] In detail, the specific implementation first decomposes the globally optimal parameters after algorithm convergence into three independent dimensions: optimal tenon standardized parameters, optimal mortise standardized parameters, and optimal proportion standardized parameters. Then, each optimal standardized parameter is multiplied by its corresponding tenon reference length, mortise reference depth, and cross-sectional reference proportion to restore the optimal tenon length, optimal mortise depth, and optimal cross-sectional proportion. Finally, the target cross-sectional area is multiplied by the optimal cross-sectional proportion, and the square root is taken to obtain the optimal equivalent width. The target cross-sectional area is divided by the optimal cross-sectional proportion, and the square root is taken to obtain the optimal equivalent height. These five optimal parameters together constitute the actual processing dimensions of the node. This method is practical. It realizes the direct mapping from digital optimization results to physical machining dimensions, without the problem of distortion from intermediate parameters. For example, the optimal tenon standardization parameter for a certain node is 1.1, the reference tenon length is 0.09 meters, and the calculated optimal tenon length is 0.099 meters. The optimal mortise standardization parameter is 1.05, the reference mortise depth is 0.10 meters, and the calculated optimal mortise depth is 0.105 meters. The optimal ratio standardization parameter is 1.0, the reference cross-section ratio is 0.8, and the calculated optimal cross-section ratio is 0.8. The target cross-sectional area is 0.0144 square meters, and the calculated optimal equivalent width is 0.1073 meters and the optimal equivalent height is 0.1342 meters.

[0172] In detail, if non-integer or numerical deviations exceeding machining accuracy occur during the optimal parameter analysis process, all length parameters are rounded to a precision of 0.001 meters to ensure the feasibility of CNC machining. The accuracy control standard for actual machining dimensions is a machining tolerance of ±0.001 meters for length parameters, with cross-sectional proportions retained to three decimal places. This precision matches the conventional process requirements for CNC machining of bamboo and wood mortise and tenon joints. The length unit in the design calculation stage is meters, which is uniformly converted to millimeters before being converted into CNC machining instructions. Numerical precision is retained without rounding during the unit conversion. When converting actual machining dimensions into CNC machining instructions, the tool center trajectory is generated based on the surface normal of the node's 3D model. If there is a conflict between the machining dimension and the tool dimension, the internal corners of the node are rounded before generating the trajectory. When there is interference between the actual machining dimension and the dimensions of the bamboo and wood component's parent material, the optimal tenon length and optimal mortise depth are adjusted proportionally while ensuring the tenon and tenon fitting allowance. The adjustment range does not exceed 5% of the original dimensions. The batch output format of the actual machining dimensions includes node identifiers, various dimensional parameters, machining accuracy requirements, and tool compatibility information. The output files are compatible with the common data formats of CNC machining equipment.

[0173] like Figure 2 As shown, Figure 2The system includes: on the left side of the screen, a 3D point cloud of the sloping site and its covering, used to extract geometric exposure features such as slope exposure and directional openness; in the middle of the screen, an integrated mortise and tenon structure building constructed on the slope, with candidate mortise and tenon joints, wooden components, and bamboo components marked on the building, while also indicating the direction of extreme lateral loads, reflecting the stress requirements of the building under extreme conditions; on the right side of the screen, a magnified view of the local design variables of the nodes, including key parameters such as mortise and tenon depth and bamboo surface ratio, and the path to directly convert the optimal node parameters through CNC machining, presenting a fully closed-loop intelligent design system from site geometric information acquisition, building structure modeling and variable optimization to final CNC machining.

[0174] It should be noted that the interval and threshold sizes are set for ease of comparison. The size of the threshold depends on the amount of sample data and the base number set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless calculations, and the formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0175] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. A smart design method for integrated mortise and tenon structure buildings based on 3D modeling, characterized in that, include: Extract the slope exposure coefficient and directional openness coefficient of the candidate mortise and tenon joints, and convert the slope exposure coefficient and directional openness coefficient into equivalent moisture content fluctuation amplitude; A three-dimensional initial model is established, and the tenon length, mortise depth and cross-sectional ratio of the candidate mortise and tenon joints in the three-dimensional initial model are set as design variables to calculate the mortise and tenon fitment margin. Extract the equivalent elastic modulus and vascular bundle distribution ratio of the components connected to the candidate mortise and tenon joints to establish the bamboo gradient amplification factor; combine the bamboo gradient amplification factor, the equivalent moisture content fluctuation amplitude and the mortise and tenon embedment margin to calculate the directional embedment attenuation. Obtain the initial stiffness of the candidate mortise and tenon joints; The initial stiffness is reduced by the directional embedment attenuation to obtain an equivalent lateral spring stiffness; The design variables are standardized to obtain standardized design parameters, and a search space is established with the overall lateral stiffness as the target. The standardized design parameters are iterated using an optimization algorithm to dynamically update the three-dimensional initial model, and the equivalent lateral spring stiffness is assembled. The overall lateral stiffness is calculated until convergence to obtain the optimal parameters. The optimal parameters are then restored to the actual machining dimensions.

2. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 1, characterized in that, Extracting the slope exposure coefficient and directional openness coefficient of candidate mortise and tenon joints, and converting the slope exposure coefficient and directional openness coefficient into equivalent moisture content fluctuation amplitudes, including: Obtain the local terrain normal unit vector, and the unit vector of the facade where the candidate mortise and tenon node is located facing the extreme lateral load direction; The local terrain normal unit vector and the unit vector of the facade where the candidate tenon and mortise node is located facing the extreme lateral load direction are multiplied by a dot product. The maximum value between the dot product result and zero is summed and the average value is calculated to obtain the directional openness coefficient. Extract the vertical component of the average normal corresponding to the candidate mortise and tenon node, and subtract the absolute value of the vertical component to obtain the slope exposure coefficient. Obtain the site reference moisture content fluctuation amplitude, extreme direction exposure coefficient, openness weight coefficient, slope weight coefficient, and windward weight coefficient; The first product is obtained by multiplying the directional openness coefficient by the openness weight coefficient, the second product is obtained by multiplying the slope exposure coefficient by the slope weight coefficient, and the third product is obtained by multiplying the extreme directional exposure coefficient by the windward weight coefficient. Summing the first product, the second product, and the third product yields the combination coefficients; The equivalent moisture content fluctuation amplitude is obtained by multiplying the site reference moisture content fluctuation amplitude by the combination coefficient.

3. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 2, characterized in that, A three-dimensional initial model is established, and the tenon length, mortise depth, and cross-sectional ratio of the candidate mortise and tenon joints in the three-dimensional initial model are set as design variables to calculate the mortise and tenon fitment margin, including: Obtain the target cross-sectional area of ​​the candidate mortise and tenon joint; Multiply the target cross-sectional area by the cross-sectional ratio, and take the square root of the multiplication result to obtain the equivalent width; Divide the target cross-sectional area by the cross-sectional ratio, and take the square root of the division result to obtain the equivalent height; Multiply the tenon length by the mortise depth and divide by the equivalent height to obtain the first calculated value; Multiply the cross-sectional ratio by two, then divide by the sum of the cross-sectional ratio and the sum of the two values ​​to obtain the second calculated value. Multiply the first calculated value by the second calculated value to obtain the tenon and mortise fitting allowance.

4. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 3, characterized in that, Extracting the equivalent elastic modulus and vascular bundle distribution ratio of the components connected to the candidate mortise and tenon joints to establish the bamboo gradient amplification factor includes: Extract the equivalent elastic modulus and the vascular bundle distribution ratio of the candidate mortise and tenon joint connected components; wherein, the equivalent elastic modulus includes the outer layer equivalent elastic modulus and the inner layer equivalent elastic modulus, and the vascular bundle distribution ratio includes the outer layer vascular bundle distribution ratio and the inner layer vascular bundle distribution ratio; Divide the equivalent elastic modulus of the outer layer by the equivalent elastic modulus of the inner layer, and perform a square root operation on the division result to obtain the first characteristic value; Divide the distribution ratio of the outer vascular bundles by the distribution ratio of the inner vascular bundles, and then take the fourth root of the division result to obtain the second characteristic value. Multiply the first feature value by the second feature value to obtain the gradient amplification coefficient of the bamboo material.

5. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 4, characterized in that, Combining the bamboo gradient amplification factor, the equivalent moisture content fluctuation amplitude, and the tenon and mortise fixation margin, the directional fixation attenuation is calculated, including: Obtain the radial shrinkage coefficient of the wood for the candidate mortise and tenon joints; The directional gap growth is obtained by multiplying the radial shrinkage coefficient of the wood, the equivalent height, and the equivalent moisture content fluctuation amplitude. Divide the directional gap growth by the tenon and mortise fitting allowance to obtain the division result; The directional embedding attenuation is obtained by multiplying the result of the phase division with the gradient amplification factor of the bamboo material.

6. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 5, characterized in that, The initial stiffness is reduced by the directional embedment attenuation to obtain an equivalent lateral spring stiffness, including: Obtain the local equivalent compression modulus, node topology correction coefficient, attenuation sensitivity coefficient, and lever arm height of the candidate mortise and tenon joint; The initial stiffness is obtained by multiplying the local equivalent compression modulus by the equivalent width, multiplying by the cube of the equivalent height, and then dividing by the product of the tenon length and twelve. Multiply the attenuation sensitivity coefficient by the directional embedding attenuation amount and take the negative value, then perform an exponential operation with the natural constant as the base to obtain the characteristic attenuation coefficient; Divide the mortise depth by the sum of the mortise depth and the tenon length to obtain the depth reduction factor; Multiply the node topology correction coefficient, the initial stiffness, the characteristic attenuation coefficient, and the depth reduction coefficient together to obtain the rotational stiffness. The equivalent lateral spring stiffness is obtained by dividing the retaining rotational stiffness by the square of the lever arm height.

7. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 6, characterized in that, The design variables are standardized to obtain standardized design parameters. A search space is established with the overall lateral stiffness as the target, including: Obtain the reference length of the tenon, the reference depth of the mortise, the reference ratio of the cross section, the maximum mortise depth, and the reference lateral stiffness of the candidate mortise and tenon joint; obtain the tool radius and machining allowance of the machining tool; Divide the tenon length by the tenon reference length to obtain the tenon standardized parameters; Divide the mortise depth by the mortise reference depth to obtain the mortise standardized parameters; Divide the cross-sectional ratio by the cross-sectional reference ratio to obtain the ratio standardization parameter; The standardized parameters of the tenon, the standardized parameters of the mortise, and the standardized parameters of the proportion together constitute the standardized design parameters; Divide the overall lateral stiffness by the reference lateral stiffness to obtain the optimization target value, and construct the search space guided by the optimization target value; Within the search space, the length of the tenon is limited to be no greater than the mortise depth, and the mortise depth is limited to be no greater than the maximum mortise depth. The minimum machining width is obtained by multiplying the tool radius by two and adding it to the machining allowance. The equivalent width is then restricted to be no less than the minimum machining width within the search space.

8. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 7, characterized in that, The standardized design parameters are iteratively evaluated using an optimization algorithm to dynamically update the three-dimensional initial model. The equivalent lateral spring stiffness is then assembled, and the overall lateral stiffness is calculated until convergence to obtain the optimal parameters. This process includes: Obtain the inertia weight, individual learning factor, group learning factor, first random number, and second random number; The difference between the individual's historical best parameter and the current generation's standardized design parameter is calculated, and the difference result, the individual's learning factor, and the first random number are multiplied together to obtain the first update term. The difference between the group's historical best parameter and the current generation's standardized design parameter is calculated, and the difference result, the group's learning factor, and the second random number are multiplied together to obtain the second update term. The inertial weight is multiplied by the current generation parameter update speed and added to the first update term and the second update term to obtain the next generation parameter update speed; The current generation of standardized design parameters is added to the update speed of the next generation of parameters to obtain the next generation of standardized design parameters, which are then used as the standardized design parameters after the iteration. Assemble the overall stiffness matrix based on the equivalent lateral spring stiffness; obtain the extreme load vector, solve for the displacement vector using the overall stiffness matrix and the extreme load vector, and extract the control layer displacement vector; The overall lateral stiffness is obtained by dividing the absolute value of the extreme load vector by the absolute value of the control layer displacement vector. Subtract the current generation's optimal target value from the previous generation's optimal target value and take the absolute value, then divide the absolute value by the previous generation's optimal target value to obtain the convergence error; When the convergence error meets the preset convergence condition, the historical optimal parameter of the group is taken as the optimal parameter.

9. The intelligent design method for integrated mortise and tenon structure buildings based on three-dimensional modeling according to claim 8, characterized in that, Restoring the optimal parameters to the actual machining dimensions includes: The optimal parameters are analyzed into the optimal tenon standardization parameter, the optimal mortise standardization parameter, and the optimal proportion standardization parameter; The optimal tenon length is obtained by multiplying the optimal tenon standardization parameter by the tenon reference length. The optimal mortise standardization parameter is multiplied by the mortise reference depth to obtain the optimal mortise depth; The optimal cross-sectional ratio is obtained by multiplying the optimal ratio standardization parameter by the cross-sectional reference ratio. Multiply the target cross-sectional area by the optimal cross-sectional ratio, and take the square root of the multiplication result to obtain the optimal equivalent width; Divide the target cross-sectional area by the optimal cross-sectional ratio, and take the square root of the division result to obtain the optimal equivalent height; The actual machining dimensions include the optimal tenon length, the optimal mortise depth, the optimal cross-sectional ratio, the optimal equivalent width, and the optimal equivalent height.

Images