Steel structure packing multi-parameter collaborative optimization system and method based on three-dimensional model

Abstract

The application discloses a steel structure packing multi-parameter collaborative optimization system and method based on a three-dimensional model, and belongs to the technical field of three-dimensional optimization of steel structure packing. Three-dimensional model data, packing behavior parameters and packing effect data of historical steel structure packing are collected to generate a feature database and sample clusters in a classified manner. Entity models of a transport carrier and a steel structure are constructed based on a three-dimensional model, a plurality of pre-arrangement simulation scenes are generated, packing behavior parameters are adjusted to maximize the packing effect and make the packing effect approximate, effective scenes are screened to construct a collaborative optimization target class, key parameters are locked and source labels are added, and parameter intervals of packing behavior indexes are extracted as collaborative optimization targets and output. The application realizes multi-parameter dynamic collaborative optimization, generates an optimal packing scheme with low transport cost, high space utilization and stable stacking, solves the problems of experience dependence and parameter collaborative deficiency in traditional packing, and improves the scientificity and transport safety of steel structure packing.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional optimization technology for steel structure packaging, specifically to a multi-parameter collaborative optimization system and method for steel structure packaging based on a three-dimensional model. Background Technology

[0002] Steel structures are widely used in buildings, bridges, industrial plants, and other fields due to their advantages such as high strength, good seismic resistance, and short construction period. In the implementation of steel structure projects, packaged transportation is a crucial link between production and installation; the rationality of the packaging plan directly affects transportation costs, component safety, and construction efficiency.

[0003] Currently, steel structure packing relies heavily on the on-site experience of staff, lacking systematic parametric analysis and scientific verification. Traditional packing methods suffer from the following problems: First, they fail to fully integrate the geometric and physical parameters of steel structure components (such as dimensions, weight, and center of gravity) with transportation constraints (such as carriage dimensions and load limits), resulting in low space utilization and high transportation costs. Second, the setting of packing behavior parameters (such as the number of stacking layers, placement angle, and fixing method) lacks coordination, often focusing only on a single objective (such as saving space) while ignoring stacking stability, easily leading to safety hazards such as component deformation and damage during transportation. Third, the lack of 3D model visualization simulation verification makes it difficult for packing schemes to adapt to different specifications of steel structures and transportation scenarios, and parameter adjustments lack a dynamic feedback mechanism, resulting in poor flexibility. Fourth, historical packing data is not effectively reused, making it impossible to continuously optimize packing schemes through data accumulation, thus hindering the improvement of packing efficiency and quality.

[0004] With the large-scale development of the steel structure industry, the requirements for precision and efficiency in packaging and transportation are increasing, and traditional experience-based packaging methods can no longer meet actual needs. Therefore, there is an urgent need for a multi-parameter collaborative optimization technology based on three-dimensional models to achieve scientific matching and dynamic adjustment of steel structure packaging parameters, taking into account transportation costs, space utilization, and stacking stability, and providing a systematic solution for steel structure packaging. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-parameter collaborative optimization system and method for steel structure packaging based on a three-dimensional model, so as to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A multi-parameter collaborative optimization system for steel structure packaging based on a 3D model is presented. The system includes: a data acquisition and processing module, a 3D pre-layout simulation module, a collaborative optimization target class construction module, and a parameter optimization feedback module.

[0008] The data acquisition and processing module is used to collect historical packaging data related to steel structures, and classify and generate a multi-dimensional feature database and corresponding feature sample clusters.

[0009] The three-dimensional pre-arrangement simulation module simulates the pre-arrangement of the steel structure based on the three-dimensional model, generates a variety of simulation scenarios, and maximizes the approximation of the packaging effect under different scenarios by adjusting the relevant parameters of the packaging behavior.

[0010] The collaborative optimization target class construction module is used to filter simulated scenarios that meet the requirements for approximating the packaging effect, construct collaborative optimization target classes, and lock the corresponding effective packaging behavior parameters.

[0011] The parameter optimization feedback module, based on the locking result, provides feedback on the effective parameter range of the packaging behavior index, uses this range as the collaborative optimization target and outputs it, thereby realizing the collaborative optimization of multiple parameters of steel structure packaging.

[0012] As a preferred embodiment of the present invention, the data acquisition and processing module includes a data acquisition unit, a feature database generation unit, and a feature sample cluster construction unit;

[0013] The data acquisition unit is used to collect historical packing data of steel structures of different specifications, including three-dimensional model data of steel structures, packing behavior parameters and packing effect data;

[0014] The feature database generation unit is used to classify the collected data and generate a first, second, and third feature database.

[0015] The feature sample cluster construction unit is used to construct corresponding three-dimensional feature sample clusters based on the classified database, ensuring that the dimension identifiers correspond one-to-one and that sample clusters of the same dimension are not repeated.

[0016] As a preferred embodiment of the present invention, the three-dimensional pre-arrangement simulation module includes a simulation scene generation unit, a packaging parameter adjustment unit, and an approximation determination unit;

[0017] The simulation scene generation unit generates several pre-arranged simulation scenes based on the coupling relationship between the 3D model and the initial packaged data.

[0018] The packaging parameter adjustment unit is used to adjust the packaging behavior parameters in the second feature sample cluster, guided by the first feature sample cluster.

[0019] The approximation determination unit is used to preset the threshold range of packaging effect difference and determine whether the packaging effect of the two scenes after adjustment meets the maximum approximation requirement.

[0020] As a preferred embodiment of the present invention, the collaborative optimization target class construction module includes an effective scene filtering unit, a target parameter locking unit, and a source tag attaching unit;

[0021] The effective scene filtering unit is used to filter out all pre-arranged simulation scenes that meet the maximum approximation requirement with the initial scene, and form a collaborative optimization target class.

[0022] The target parameter locking unit is used to lock the second feature sample cluster that meets the requirements as the adjusted target sample cluster.

[0023] The source label attaching unit is used to attach source labels to the second feature sample clusters of the initial scene and the effective scene, respectively.

[0024] As a preferred embodiment of the present invention, the parameter optimization feedback module includes a parameter range extraction unit and an optimization target determination unit;

[0025] The parameter range extraction unit extracts the parameter range of each packaging behavior indicator based on the source label of the second feature sample cluster in the effective scenario.

[0026] The optimization target determination unit is used to take the extracted parameter range as the collaborative optimization target corresponding to the initial scene and output the target.

[0027] A multi-parameter collaborative optimization method for steel structure packaging based on a 3D model, comprising the following steps:

[0028] Step S1: Collect relevant historical packaging data of steel structures, and classify and generate a multi-dimensional feature database and corresponding feature sample clusters;

[0029] Step S2: Based on the 3D model, perform pre-arrangement simulation of the steel structure to generate multiple simulation scenarios. By adjusting the relevant parameters of the packing behavior, maximize the approximation of the packing effect under different scenarios.

[0030] Step S3: Filter the simulation scenarios that meet the requirements for approximating the packaging effect, construct the collaborative optimization target class, and lock the corresponding effective packaging behavior parameters;

[0031] Step S4: Based on the locking results, the effective parameter range of the packing behavior index is fed back, and this range is used as the collaborative optimization target and output to realize the collaborative optimization of multiple parameters of steel structure packing.

[0032] As a preferred embodiment of the present invention, the specific implementation process of step S1 includes:

[0033] Historical packing data of steel structures of different specifications were collected, including three-dimensional model data of steel structures, packing behavior parameters, and packing effect data, in order to generate a three-dimensional feature database and a three-dimensional feature sample cluster respectively.

[0034] The three-dimensional feature database includes a first feature database containing only three-dimensional model data of steel structures, a second feature database containing only packaging behavior parameters, and a third feature database containing only packaging effect data.

[0035] The three-dimensional feature sample clusters are the first feature sample cluster, the second feature sample cluster, and the third feature sample cluster.

[0036] The dimensions of the three-dimensional feature database and the three-dimensional feature sample clusters are bound to each other in a one-to-one dimension identifier relationship. Each dimension feature database contains several feature sample clusters of that dimension, and each dimension feature sample cluster is composed of packaged data of the same type under that dimension. Furthermore, the feature sample clusters under the same dimension are all different.

[0037] As a preferred embodiment of the present invention, the specific implementation process of step S2 includes:

[0038] Based on the three-dimensional model, the steel structure is pre-arranged and simulated. In the three-dimensional model space of the transport vehicle, according to the coupling relationship of the initial packing data, several pre-arrangement simulation scenarios are generated. Each pre-arrangement simulation scenario is composed of a feature sample cluster corresponding to different dimensions in series. The coupling relationship is expressed as: Pre-arrangement simulation scenario: first feature sample cluster → second feature sample cluster → third feature sample cluster.

[0039] Based on the first feature sample cluster, when packaging steel structures of the same specification, the same first feature sample cluster is used as a guide. Addressing the differences between the second and third feature sample clusters, the packaging behavior parameters in the second feature sample cluster are adjusted to maximize the approximation between the second feature sample cluster before and after adjustment and the corresponding third feature sample cluster. The maximization approximation process is as follows:

[0040] The i-th coupling relationship is represented as follows: pre-arranged simulation scenario i: first feature sample cluster X1 → second feature sample cluster X2 → third feature sample cluster X3;

[0041] The j-th coupling relationship can be represented as: pre-arranged simulation scenario j: first feature sample cluster X1 → second feature sample cluster →Third Feature Sample Cluster And i≠j;

[0042] Adjust the second feature sample cluster X2 to the second feature sample cluster. Then, the third feature sample cluster X3 and the third feature sample cluster are evaluated. The degree of approximation of the difference parameters between them, that is, the degree of difference of the second parameter between pre-arranged simulation scene i and pre-arranged simulation scene j. In the formula, This represents the type parameter of the e-th packaging effect indicator. This indicates the type of parameter representing the packaging effect index derived from the third feature sample cluster X3. This indicates that the sample originates from the third feature cluster. The packaging performance indicator type parameter, where E represents the total number of packaging performance indicator types;

[0043] The second parameter, the degree of difference, is preset within a threshold range. If the second parameter, the degree of difference... If the sample falls within the second parameter's difference threshold range, then the third feature sample cluster X3 is determined to be different from the third feature sample cluster X4. If the maximum approximation requirement is met, then the third feature sample cluster X3 is determined to be similar to the third feature sample cluster X4. The maximum approximation requirement is not met.

[0044] As a preferred embodiment of the present invention, the specific implementation process of step S3 includes:

[0045] Obtain all pre-arranged simulation scenarios that satisfy the maximization approximation requirement with respect to pre-arranged simulation scenario i, form a collaborative optimization target class, and lock the second feature sample cluster that satisfies the maximization approximation requirement. , to be used as the second feature sample cluster X2 after adjustment;

[0046] Based on the locking results, add source labels to the second feature sample cluster X2 in the pre-arranged simulation scenario i. ,and The second feature sample cluster in the pre-arranged simulation scenario j that satisfies the maximum approximation requirement with the pre-arranged simulation scenario i. Additional source tags ,and .

[0047] As a preferred embodiment of the present invention, the specific implementation process of step S4 includes:

[0048] Source tags for pre-arranged simulation scenario j Feedback on packaging behavior indicators and parameter ranges that affect the achievement of packaging standards In the formula, This represents the parameter type for the r-th packaging behavior. Indicates that it originates from the source tag. The packaging behavior index type parameter J represents the set of collaborative optimization objectives consisting of all pre-arranged simulation scenarios that satisfy the maximization approximation requirement with respect to pre-arranged simulation scenario i. This is a minimum value feedback indicator function, used to select the minimum packaging behavior indicator type parameter. This is a maximum value feedback indicator function, used to select the parameter of the highest packaging behavior indicator type. For the range of parameters of the r-th packaging behavior indicator type;

[0049] Source tags based on pre-arranged simulation scenario i The packaged behavior indicator type parameter range As a source tag The collaborative optimization objective is determined by the type parameters of the packaging behavior indicators, and the collaborative optimization objective is output.

[0050] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: By collecting 3D model data, packing behavior parameters, and packing effect data of historical steel structure packing, a feature database and sample clusters are generated and categorized; based on the 3D model, a physical model of the transport carrier and the steel structure is constructed, generating various pre-arrangement simulation scenarios, and adjusting the packing behavior parameters to maximize the packing effect; effective scenarios are selected to construct a collaborative optimization target class, key parameters are locked and source tags are attached; the parameter range of the packing behavior indicators is extracted as the collaborative optimization target and output. This invention achieves multi-parameter dynamic collaborative optimization, generating an optimal packing scheme with low transportation costs, high space utilization, and stable stacking, solving the problems of traditional packing relying on experience and insufficient parameter coordination, and improving the scientific nature and transportation safety of steel structure packing. Attached Figure Description

[0051] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0052] Figure 1 This is a schematic diagram illustrating the steps of the multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model according to the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] In this first embodiment: a multi-parameter collaborative optimization system for steel structure packaging based on a three-dimensional model is provided. The system includes: a data acquisition and processing module, a three-dimensional pre-arrangement simulation module, a collaborative optimization target class construction module, and a parameter optimization feedback module.

[0055] The data acquisition and processing module is used to collect historical packaging data related to steel structures, and classify and generate a multi-dimensional feature database and corresponding feature sample clusters.

[0056] The data acquisition and processing module includes a data acquisition unit, a feature database generation unit, and a feature sample cluster construction unit.

[0057] The data acquisition unit is used to collect historical packing data of steel structures of different specifications, including three-dimensional model data of steel structures, packing behavior parameters and packing effect data;

[0058] The feature database generation unit is used to classify the collected data and generate first, second, and third feature databases.

[0059] The feature sample cluster construction unit is used to construct corresponding three-dimensional feature sample clusters based on the classified database, ensuring that the dimension labels correspond one-to-one and that sample clusters of the same dimension are not repeated.

[0060] The 3D pre-layout simulation module simulates the pre-layout of the steel structure based on the 3D model, generates a variety of simulation scenarios, and maximizes the approximation of the packaging effect under different scenarios by adjusting the relevant parameters of the packaging behavior.

[0061] The 3D pre-arrangement simulation module includes a simulation scene generation unit, a packaging parameter adjustment unit, and an approximation determination unit.

[0062] The simulation scene generation unit generates several pre-arranged simulation scenes based on the coupling relationship between the 3D model and the initial packaged data;

[0063] The packing parameter adjustment unit is used to adjust the packing behavior parameters in the second feature sample cluster, guided by the first feature sample cluster.

[0064] The approximation judgment unit is used to preset the threshold range of the difference in packaging effect and determine whether the packaging effect of the two scenes after adjustment meets the maximum approximation requirement.

[0065] The collaborative optimization target class construction module is used to filter simulation scenarios that meet the requirements for approximating the packaging effect, construct collaborative optimization target classes, and lock the corresponding effective packaging behavior parameters;

[0066] The collaborative optimization target class construction module includes an effective scenario filtering unit, a target parameter locking unit, and a source tag attachment unit.

[0067] The effective scene filtering unit is used to filter out all pre-arranged simulation scenes that meet the maximum approximation requirement with the initial scene, forming a collaborative optimization target class;

[0068] The target parameter locking unit is used to lock the second feature sample cluster that meets the requirements, and use it as the adjusted target sample cluster.

[0069] The source label appending unit is used to append source labels to the second feature sample clusters of the initial scene and the effective scene, respectively;

[0070] The parameter optimization feedback module, based on the locking result, provides feedback on the effective parameter range of the packing behavior index, uses this range as the collaborative optimization target and outputs it, thereby realizing the collaborative optimization of multiple parameters of steel structure packing;

[0071] The parameter optimization feedback module includes a parameter range extraction unit and an optimization target determination unit.

[0072] The parameter range extraction unit extracts the parameter range of each packaging behavior indicator based on the source label of the second feature sample cluster in the effective scenario.

[0073] The optimization target determination unit is used to take the extracted parameter range as the collaborative optimization target corresponding to the initial scene and output the target.

[0074] Please see Figure 1 In this second embodiment, a multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model is provided to be applicable to the first embodiment. This embodiment takes a steel structure company undertaking an industrial plant construction project as the implementation scenario. A batch of steel structure components of different specifications need to be transported, including H-beams (400×200×8×12mm, single length 12m, weight 62kg / m), I-beams (36a type, single length 10m, weight 60.03kg / m), and box columns (500×500×12×12mm, single length 8m, weight 141.3kg / m). The transport vehicle is a heavy-duty semi-trailer (cargo box size: 13m×2.5m×2.8m, load limit 30 tons). The packaging materials are 10mm thick plywood and Q235 steel strip (width 30mm, thickness 1.5mm). The packaging scheme is required to meet the targets of space utilization ≥80% and stacking tilt ≤3°.

[0075] The method includes the following steps:

[0076] Step S1: Collect relevant historical packaging data of steel structures, and classify and generate a multi-dimensional feature database and corresponding feature sample clusters;

[0077] For example, historical packing data of steel structures of different specifications are collected, including three-dimensional model data of steel structures, packing behavior parameters, and packing effect data, to generate a three-dimensional feature database and a three-dimensional feature sample cluster, respectively.

[0078] The three-dimensional feature database includes a first feature database containing only three-dimensional model data of steel structures, a second feature database containing only packing behavior parameters, and a third feature database containing only packing effect data.

[0079] The three-dimensional feature sample clusters are the first feature sample cluster, the second feature sample cluster, and the third feature sample cluster.

[0080] The dimensions of the three-dimensional feature database and the three-dimensional feature sample clusters are bound to each other in a one-to-one dimension identifier relationship. Each dimension feature database contains several feature sample clusters of that dimension, and each feature sample cluster is composed of packaged data of the same type under that dimension. Furthermore, the feature sample clusters under the same dimension are all different.

[0081] For example, by using the enterprise's ERP system and historical transportation records, 50 sets of packaging data for similar steel structures were collected, including:

[0082] 3D model data: length, width, height, center of gravity coordinates (center of gravity coordinates of H-beams are ±2mm from the center of the web, and center of gravity coordinates of I-beams are ±1.5mm from the center of the flange), cross-sectional dimension tolerances, etc. for each component;

[0083] Packaging parameters: number of stacking layers (2-5 layers), placement angle (0°-90°), packaging material thickness (8-12mm), steel strap fixing spacing (0.8-1.5m), support component placement, etc.

[0084] Packaging performance data: space utilization rate (65%-78%), transportation damage rate (1.2%-3.5%), unit transportation cost (120-180 yuan / ton), stacking tilt angle (2.5°-4.8°), etc.

[0085] The collected data is classified into three feature databases: the first feature database contains the 3D model parameters of all components, the second feature database contains various packaging behavior parameters, and the third feature database contains the corresponding packaging effect indicators.

[0086] Based on the differences in data types, we construct a first feature sample cluster (divided into three sub-clusters according to component type: H-beams, I-beams, and box columns), a second feature sample cluster (divided into 12 sub-clusters according to stacking layers and placement angle combinations), and a third feature sample cluster (divided into 3 sub-clusters according to space utilization level), ensuring that sample clusters under the same dimension are not repeated and that dimension identifiers correspond one-to-one.

[0087] Step S2: Based on the 3D model, perform pre-arrangement simulation of the steel structure to generate multiple simulation scenarios. By adjusting the relevant parameters of the packing behavior, maximize the approximation of the packing effect under different scenarios.

[0088] For example, based on the three-dimensional model, the steel structure is pre-arranged and simulated. In the three-dimensional model space of the transport vehicle, according to the coupling relationship of the initial packing data, several pre-arrangement simulation scenarios are generated. Each pre-arrangement simulation scenario is composed of a feature sample cluster corresponding to different dimensions in series. The coupling relationship is represented as: pre-arrangement simulation scenario: first feature sample cluster → second feature sample cluster → third feature sample cluster.

[0089] Based on the first feature sample cluster, when packing steel structures of the same specification, the same first feature sample cluster is used as a guide. Addressing the differences between the second and third feature sample clusters, the packing behavior parameters in the second feature sample cluster are adjusted to maximize the approximation between the second feature sample cluster before and after adjustment and the corresponding third feature sample cluster. The maximization approximation process is as follows:

[0090] The i-th coupling relationship is represented as follows: pre-arranged simulation scenario i: first feature sample cluster X1 → second feature sample cluster X2 → third feature sample cluster X3;

[0091] The j-th coupling relationship can be represented as: pre-arranged simulation scenario j: first feature sample cluster X1 → second feature sample cluster →Third Feature Sample Cluster And i≠j;

[0092] Adjust the second feature sample cluster X2 to the second feature sample cluster. Then, the third feature sample cluster X3 and the third feature sample cluster are evaluated. The degree of approximation of the difference parameters between them, that is, the degree of difference of the second parameter between pre-arranged simulation scene i and pre-arranged simulation scene j. In the formula, This represents the type parameter of the e-th packaging effect indicator. This indicates the type of parameter representing the packaging effect index derived from the third feature sample cluster X3. This indicates that the sample originates from the third feature cluster. The packaging performance indicator type parameter, where E represents the total number of packaging performance indicator types;

[0093] The second parameter, the degree of difference, is preset within a threshold range. If the second parameter, the degree of difference... If the sample falls within the second parameter's difference threshold range, then the third feature sample cluster X3 is determined to be different from the third feature sample cluster X4. If the maximum approximation requirement is met, then the third feature sample cluster X3 is determined to be similar to the third feature sample cluster X4. The maximum approximation requirement is not met;

[0094] For example, in Revit software, a 3D solid model of the transport vehicle compartment and various steel structural components is constructed. According to the coupling relationship of the initial packing data (first feature sample cluster → second feature sample cluster → third feature sample cluster), 20 pre-arrangement simulation scenarios are generated. For example, scenario 1: H-beam (first feature sub-cluster 1) → 3-layer stacking + 0° placement (second feature sub-cluster 4) → space utilization rate 72% (third feature sub-cluster 2); scenario 8: I-beam (first feature sub-cluster 2) → 4-layer stacking + 15° placement (second feature sub-cluster 7) → space utilization rate 76% (third feature sub-cluster 2), etc.

[0095] Guided by the first feature sub-cluster 1 of the H-beams, the packaging behavior parameters are adjusted according to the differences in the third feature sub-clusters corresponding to different second feature sub-clusters. For example, in the initial scenario, the number of H-beam stacks is 4 layers, and the total weight is 4 layers × 20 pieces × 744 kg / piece = 59.52 tons, which exceeds the vehicle's 30-ton load limit. The number of stacks is adjusted to 3 layers, and the total weight is 3 layers × 16 pieces × 744 kg / piece = 35.712 tons, which still exceeds the limit. The weight is then adjusted to 3 layers × 13 pieces × 744 kg / piece = 28.992 tons, which meets the load constraint. At the same time, when the initial placement angle is 0°, the center of gravity offset of the stack is 12 cm and the tilt is 3.8°. When the placement angle is adjusted to 15°, the center of gravity offset is reduced to 8 cm and the tilt is 2.6°, which meets the constraint of ≤3°.

[0096] The preset threshold range for the difference in packaging effect is [0.1, 0.3]. By calculating the comprehensive difference in packaging effect (space utilization, tilt, cost) before and after adjustment, 12 effective simulation scenarios that meet the threshold requirements are selected. For example, the difference between scenario 1 after adjustment (3 layers of stacking + 15° placement) and scenario 6 (3 layers of stacking + 20° placement) is 0.22, which is within the threshold range, and it is determined that the maximum approximation requirement is met.

[0097] Step S3: Filter the simulation scenarios that meet the requirements for approximating the packaging effect, construct the collaborative optimization target class, and lock the corresponding effective packaging behavior parameters;

[0098] For example, all pre-arranged simulation scenarios that satisfy the maximum approximation requirement with respect to the pre-arranged simulation scenario i are obtained, forming a collaborative optimization target class, and the second feature sample cluster that satisfies the maximum approximation requirement is locked. , to be used as the second feature sample cluster X2 after adjustment;

[0099] Based on the locking results, add source labels to the second feature sample cluster X2 in the pre-arranged simulation scenario i. ,and The second feature sample cluster in the pre-arranged simulation scenario j that satisfies the maximum approximation requirement with the pre-arranged simulation scenario i. Additional source tags ,and ;

[0100] For example, from 20 pre-arranged simulation scenarios, 12 scenarios that meet the maximum approximation requirement with the initial scenario (3-layer stacking of H-beams + 0° placement, 3-layer stacking of I-beams + 0° placement, 2-layer stacking of box columns + 0° placement) are selected to form a collaborative optimization target class;

[0101] Lock the second feature sample cluster corresponding to the target class, namely the combination of packaging behavior parameters: H-beams (3-layer stacking, 15° placement, steel strip spacing 1.2m), I-beams (3-layer stacking, 10° placement, steel strip spacing 1.0m), box columns (2-layer stacking, 0° placement, with added wooden support components);

[0102] The initial scene's second feature sample cluster is labeled with "X2(0)", and the 12 valid scene's second feature sample clusters are labeled with "X2(1)" to "X2(12)" respectively, clarifying the source correspondence of each parameter, which facilitates subsequent parameter tracing and adjustment.

[0103] Step S4: Based on the locking results, the effective parameter range of the packing behavior index is fed back, and this range is used as the collaborative optimization target and output to achieve collaborative optimization of multiple parameters of steel structure packing;

[0104] For example, based on the source labels of the pre-arranged simulation scenario j Feedback on packaging behavior indicators and parameter ranges that affect the achievement of packaging standards In the formula, This represents the parameter type for the r-th packaging behavior. Indicates that it originates from the source tag. The packaging behavior index type parameter J represents the set of collaborative optimization objectives consisting of all pre-arranged simulation scenarios that satisfy the maximization approximation requirement with respect to pre-arranged simulation scenario i. This is a minimum value feedback indicator function, used to select the minimum packaging behavior indicator type parameter. This is a maximum value feedback indicator function, used to select the parameter of the highest packaging behavior indicator type. For the range of parameters of the r-th packaging behavior indicator type;

[0105] Source tags based on pre-arranged simulation scenario i The packaged behavior indicator type parameter range As a source tag The collaborative optimization objective is determined by the type parameters of the packaging behavior indicators, and the collaborative optimization objective is output.

[0106] For example, parameter range extraction: Based on the packaging behavior parameters corresponding to the labels “X2(1)” to “X2(12)”, the parameter ranges of each packaging behavior indicator are extracted:

[0107] Stacking layers: 2-3 layers for H-beams, 2-3 layers for I-beams, and 1-2 layers for box columns;

[0108] Placement angles: H-beams 10°-20°, I-beams 8°-15°, box columns 0°-5°;

[0109] Steel strip spacing: 0.9m-1.3m;

[0110] Packaging material thickness: 9mm-11mm;

[0111] Using the above parameter range as the initial collaborative optimization objective, we substitute it into the multi-objective optimization model (aiming at the lowest transportation cost, highest space utilization, and strongest stacking stability) for final calculation to generate the optimal packaging scheme:

[0112] H-beams: stacked in 3 layers at a 15° angle, with steel strip spacing of 1.2m, and packaged with 10mm thick plywood;

[0113] I-beams: stacked in 3 layers, placed at a 12° angle, with steel strip spacing of 1.0m, and packaged with 10mm thick plywood;

[0114] Box-shaped columns: 2 layers stacked, placed at a 3° angle, with 2 additional wooden supports, and packaged with 10mm thick plywood.

[0115] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0116] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model, characterized in that, The method includes the following steps: Step S1: Collect relevant historical packaging data of steel structures, and classify and generate a multi-dimensional feature database and corresponding feature sample clusters; Step S2: Based on the 3D model, perform pre-arrangement simulation of the steel structure to generate multiple simulation scenarios. By adjusting the relevant parameters of the packing behavior, maximize the approximation of the packing effect under different scenarios. Step S3: Filter the simulation scenarios that meet the requirements for approximating the packaging effect, construct the collaborative optimization target class, and lock the corresponding effective packaging behavior parameters; Step S4: Based on the locking results, the effective parameter range of the packing behavior index is fed back, and this range is used as the collaborative optimization target and output to achieve collaborative optimization of multiple parameters of steel structure packing; The specific implementation process of step S2 includes: Based on the three-dimensional model, the steel structure is pre-arranged and simulated. In the three-dimensional model space of the transport vehicle, according to the coupling relationship of the initial packing data, several pre-arrangement simulation scenarios are generated. Each pre-arrangement simulation scenario is composed of a feature sample cluster corresponding to different dimensions in series. The coupling relationship is expressed as: Pre-arrangement simulation scenario: first feature sample cluster → second feature sample cluster → third feature sample cluster. Based on the first feature sample cluster, when packaging steel structures of the same specification, the same first feature sample cluster is used as a guide. Addressing the differences between the second and third feature sample clusters, the packaging behavior parameters in the second feature sample cluster are adjusted to maximize the approximation between the second feature sample cluster before and after adjustment and the corresponding third feature sample cluster. The maximization approximation process is as follows: The i-th coupling relationship is represented as follows: pre-arranged simulation scenario i: first feature sample cluster X1 → second feature sample cluster X2 → third feature sample cluster X3; The j-th coupling relationship can be represented as: pre-arranged simulation scenario j: first feature sample cluster X1 → second feature sample cluster Third feature sample cluster And i≠j; Adjust the second feature sample cluster X2 to the second feature sample cluster. Then, the third feature sample cluster X3 and the third feature sample cluster are evaluated. The degree of approximation of the difference parameters between them, that is, the degree of difference of the second parameter between pre-arranged simulation scene i and pre-arranged simulation scene j. In the formula, This represents the type parameter of the e-th packaging effect indicator. This indicates the type of parameter representing the packaging effect index derived from the third feature sample cluster X3. This indicates that the sample originates from the third feature cluster. The packaging performance indicator type parameter, where E represents the total number of packaging performance indicator types; The second parameter, the degree of difference, is preset within a threshold range. If the second parameter, the degree of difference... If the sample falls within the second parameter's difference threshold range, then the third feature sample cluster X3 is determined to be different from the third feature sample cluster X4. If the maximum approximation requirement is met, then the third feature sample cluster X3 is determined to be similar to the third feature sample cluster X4. The maximum approximation requirement is not met.

2. The multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model according to claim 1, characterized in that, The specific implementation process of step S1 includes: Historical packing data of steel structures of different specifications were collected, including three-dimensional model data of steel structures, packing behavior parameters, and packing effect data, in order to generate a three-dimensional feature database and a three-dimensional feature sample cluster respectively. The three-dimensional feature database includes a first feature database containing only three-dimensional model data of steel structures, a second feature database containing only packaging behavior parameters, and a third feature database containing only packaging effect data. The three-dimensional feature sample clusters are the first feature sample cluster, the second feature sample cluster, and the third feature sample cluster. The dimensions of the three-dimensional feature database and the three-dimensional feature sample clusters are bound to each other in a one-to-one dimension identifier relationship. Each dimension feature database contains several feature sample clusters of that dimension, and each dimension feature sample cluster is composed of packaged data of the same type under that dimension. Furthermore, the feature sample clusters under the same dimension are all different.

3. The multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model according to claim 1, characterized in that, The specific implementation process of step S3 includes: Obtain all pre-arranged simulation scenarios that satisfy the maximization approximation requirement with respect to pre-arranged simulation scenario i, form a collaborative optimization target class, and lock the second feature sample cluster that satisfies the maximization approximation requirement. , to be used as the second feature sample cluster X2 after adjustment; Based on the locking results, add source labels to the second feature sample cluster X2 in the pre-arranged simulation scenario i. ,and The second feature sample cluster in the pre-arranged simulation scenario j that satisfies the maximum approximation requirement with the pre-arranged simulation scenario i. Additional source tags ,and .

4. The multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model according to claim 1, characterized in that, The specific implementation process of step S4 includes: Source tags for pre-arranged simulation scenario j Feedback on packaging behavior indicators and parameter ranges that affect the achievement of packaging standards In the formula, This represents the parameter type for the r-th packaging behavior. Indicates that it originates from the source tag. The packaging behavior index type parameter J represents the set of collaborative optimization objectives consisting of all pre-arranged simulation scenarios that satisfy the maximization approximation requirement with respect to pre-arranged simulation scenario i. This is a minimum value feedback indicator function, used to select the minimum packaging behavior indicator type parameter. This is a maximum value feedback indicator function, used to select the parameter of the highest packaging behavior indicator type. For the range of parameters of the r-th packaging behavior indicator type; Source tags based on pre-arranged simulation scenario i The packaged behavior indicator type parameter range As a source tag The collaborative optimization objective is determined by the type parameters of the packaging behavior indicators, and the collaborative optimization objective is output.

5. A multi-parameter collaborative optimization system for steel structure packaging based on a three-dimensional model, executing the multi-parameter collaborative optimization method for steel structure packaging based on a three-dimensional model as described in any one of claims 1-4, characterized in that, The system includes: a data acquisition and processing module, a three-dimensional pre-arrangement simulation module, a collaborative optimization target class construction module, and a parameter optimization feedback module; The data acquisition and processing module is used to collect historical packaging data related to steel structures, and classify and generate a multi-dimensional feature database and corresponding feature sample clusters. The three-dimensional pre-arrangement simulation module simulates the pre-arrangement of the steel structure based on the three-dimensional model, generates a variety of simulation scenarios, and maximizes the approximation of the packaging effect under different scenarios by adjusting the relevant parameters of the packaging behavior. The collaborative optimization target class construction module is used to filter simulated scenarios that meet the requirements for approximating the packaging effect, construct collaborative optimization target classes, and lock the corresponding effective packaging behavior parameters. The parameter optimization feedback module, based on the locking result, provides feedback on the effective parameter range of the packaging behavior index, uses this range as the collaborative optimization target and outputs it, thereby realizing the collaborative optimization of multiple parameters of steel structure packaging.

6. The multi-parameter collaborative optimization system for steel structure packaging based on a three-dimensional model according to claim 5, characterized in that, The data acquisition and processing module includes a data acquisition unit, a feature database generation unit, and a feature sample cluster construction unit; The data acquisition unit is used to collect historical packing data of steel structures of different specifications, including three-dimensional model data of steel structures, packing behavior parameters and packing effect data; The feature database generation unit is used to classify the collected data and generate a first, second, and third feature database. The feature sample cluster construction unit is used to construct corresponding three-dimensional feature sample clusters based on the classified database, ensuring that the dimension identifiers correspond one-to-one and that sample clusters of the same dimension are not repeated.

7. The multi-parameter collaborative optimization system for steel structure packaging based on a three-dimensional model according to claim 5, characterized in that, The three-dimensional pre-arrangement simulation module includes a simulation scene generation unit, a packaging parameter adjustment unit, and an approximation determination unit; The simulation scene generation unit generates several pre-arranged simulation scenes based on the coupling relationship between the 3D model and the initial packaged data. The packaging parameter adjustment unit is used to adjust the packaging behavior parameters in the second feature sample cluster, guided by the first feature sample cluster. The approximation determination unit is used to preset the threshold range of packaging effect difference and determine whether the packaging effect of the two scenes after adjustment meets the maximum approximation requirement.

8. The multi-parameter collaborative optimization system for steel structure packaging based on a three-dimensional model according to claim 5, characterized in that, The collaborative optimization target class construction module includes an effective scene filtering unit, a target parameter locking unit, and a source label attachment unit; The effective scene filtering unit is used to filter out all pre-arranged simulation scenes that meet the maximum approximation requirement with the initial scene, and form a collaborative optimization target class. The target parameter locking unit is used to lock the second feature sample cluster that meets the requirements as the adjusted target sample cluster. The source label attaching unit is used to attach source labels to the second feature sample clusters of the initial scene and the effective scene, respectively.

9. The multi-parameter collaborative optimization system for steel structure packaging based on a three-dimensional model according to claim 5, characterized in that, The parameter optimization feedback module includes a parameter range extraction unit and an optimization target determination unit; The parameter range extraction unit extracts the parameter range of each packaging behavior indicator based on the source label of the second feature sample cluster in the effective scenario. The optimization target determination unit is used to take the extracted parameter range as the collaborative optimization target corresponding to the initial scene and output the target.

Images