Shipyard steel bent frame scheme automatic optimization method and system based on genetic algorithm

Abstract

The invention relates to the technical field of data processing, and discloses a shipyard steel framed bent scheme automatic optimization method and system based on a genetic algorithm. The method comprises the following steps: acquiring shipyard steel bent frame structure parameters and establishing an engineering experience database to obtain initial component section parameters; carrying out structure parametric modeling based on the initial parameters, generating a structure calculation model and carrying out load condition analysis; importing finite element software to carry out internal force analysis, and carrying out component standard checking calculation to obtain a stress ratio; performing multi-objective optimization through an NSGA-II algorithm to obtain a Pareto optimal solution set; and finally carrying out three-dimensional modeling and document generation to obtain an optimization design result. According to the method, the design scheme of the steel bent frame structure is automatically optimized while multiple load working conditions and design targets are considered, and it is ensured that the design result meeting the safety standard and with the optimal cost is obtained within the shortest time. By solving the problem, the design efficiency can be greatly improved, the design cost is reduced, and the structural safety is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a shipyard steel rack scheme automatic optimization method and system based on a genetic algorithm. BACKGROUND

[0002] The existing shipyard steel rack design generally relies on traditional manual calculation and empirical data. This design method not only takes time and effort, but also has a certain risk of error. The design of steel rack structure usually needs to consider many complex factors, such as the strength of steel, the geometric size of components, load conditions and construction conditions, etc. At present, many designers use common manual calculation methods to select the size of components and analyze the stress, but these methods often rely on the experience and intuition of designers, and it is difficult to consider the optimization requirements under multiple working conditions. With the expansion of shipyard construction scale and the increase of steel structure design complexity, the traditional design method gradually exposes its limitations, especially in complex load conditions and multi-objective optimization, the existing method has been difficult to meet the accuracy and efficiency requirements of modern engineering design.

[0003] The deficiencies of the prior art mainly manifest in two aspects. First, the traditional design method is difficult to realize rapid and accurate multi-objective optimization. When considering the strength, stability, material consumption and construction cost of steel rack structure and other factors, designers often need to weigh between multiple schemes, and it is difficult to comprehensively evaluate all possible design options. Second, the existing technology often ignores the automation and intelligentization in the design process, especially in the aspects of component cross-section size, load condition and material selection, many design steps still rely on manual input and calculation, which makes the design period longer and is easily affected by human factors, resulting in low accuracy and reliability of the design result. Therefore, there is an urgent need for a new method that can automatically optimize the design, reduce human interference, and effectively handle multi-objective optimization problems. SUMMARY

[0004] The present application provides a shipyard steel rack scheme automatic optimization method and system based on a genetic algorithm, which is used to automatically optimize the design scheme of steel rack structure while considering multiple load conditions and design objectives, to ensure that the design result meets the safety standards and has the optimal cost in the shortest time. The solution of this problem can greatly improve the design efficiency, reduce the design cost, and ensure the structural safety.

[0005] In a first aspect, the application provides a shipyard steel truss scheme automatic optimization method based on a genetic algorithm, which comprises: collecting shipyard steel truss structure parameters, establishing an engineering experience database, and obtaining initial component cross-section parameters; performing parameterized modeling on the steel truss structure based on the initial component cross-section parameters, generating structure attributes, and generating various load cases according to specifications to obtain a structure calculation model; importing the structure calculation model into finite element software to perform internal force analysis and obtain component internal force data; performing specification checking on each component according to the component internal force data, calculating various stress ratios, and obtaining component control stress ratios; setting decision variables and constraint conditions based on the component control stress ratios, performing multi-objective optimization iteration through an NSGA-II algorithm, and obtaining a Pareto optimal solution set; and performing three-dimensional modeling and document generation on the Pareto optimal solution set to obtain final optimization design results.

[0006] In a second aspect, the application provides a shipyard steel truss scheme automatic optimization system based on a genetic algorithm, which comprises: A collection module for collecting shipyard steel truss structure parameters, establishing an engineering experience database, and obtaining initial component cross-section parameters; A modeling module for performing parameterized modeling on the steel truss structure based on the initial component cross-section parameters, generating structure attributes, and generating various load cases according to specifications to obtain a structure calculation model; An analysis module for importing the structure calculation model into finite element software to perform internal force analysis and obtain component internal force data; A checking module for performing specification checking on each component according to the component internal force data, calculating various stress ratios, and obtaining component control stress ratios; An iteration module for setting decision variables and constraint conditions based on the component control stress ratios, performing multi-objective optimization iteration through an NSGA-II algorithm, and obtaining a Pareto optimal solution set; A generation module for performing three-dimensional modeling and document generation on the Pareto optimal solution set to obtain final optimization design results.

[0007] In a third aspect, a shipyard steel truss scheme automatic optimization device based on a genetic algorithm is provided, which comprises a memory and at least one processor, the memory storing instructions; the at least one processor invokes the instructions in the memory to enable the shipyard steel truss scheme automatic optimization device based on a genetic algorithm to perform the shipyard steel truss scheme automatic optimization method based on a genetic algorithm described above.

[0008] In a fourth aspect, a computer readable storage medium is provided, in which instructions are stored, when executed on a computer, cause the computer to perform the genetic algorithm-based automatic optimization method for shipyard steel grillage scheme.

[0009] In the technical scheme provided in the present application, the genetic algorithm-based automatic optimization method for shipyard steel grillage scheme realizes efficient, accurate and automatic steel grillage structure design. Firstly, the present application ensures the rationality of the initial component cross-section parameters by collecting the shipyard steel grillage structure parameters and establishing an engineering experience database. This method breaks through the limitations of traditional design relying on experience, can quickly obtain historical design data and provide a basis for new design schemes. By combining the initial component cross-section parameters with the parametric modeling of the steel grillage structure, the system can accurately generate the geometric model of the structure and automatically generate various load conditions, making the entire design process more systematic and efficient. Compared with traditional methods, this parametric modeling not only reduces the error rate of manual operation, but also makes the structural design response more accurate under different conditions. The present application introduces a multi-objective optimization algorithm, NSGA-II genetic algorithm, in the design. This algorithm is particularly advantageous in multi-objective optimization, can simultaneously optimize multiple objectives such as the weight, component surface area and stress ratio of the steel grillage structure, and under the premise of ensuring the safety of the structure, reduces the amount of material used and the cost. The genetic algorithm uses adaptive crossover rate and mutation rate strategies to make the optimization process more flexible, avoiding the problem that traditional optimization methods may fall into local optimal solutions. By using the multi-objective optimization characteristics of the genetic algorithm, the present application can balance between different design schemes and help designers select the most cost-effective design scheme. At the same time, the algorithm can screen the population through non-dominated sorting and crowding degree calculation, ensuring the efficiency and diversity of the optimization process and ensuring that the final Pareto optimal solution set has the best performance. By introducing an artificial intelligence model to perform data mining and pattern recognition on historical design cases, the system can quickly find the most similar historical case from the database according to new input parameters and extract the initial component cross-section parameters, providing a reasonable starting point for subsequent parametric modeling and optimization iteration. This intelligent data matching process greatly improves design efficiency, avoids errors and inconsistencies caused by manual judgment, and ensures the scientificity and rationality of the design scheme. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative labor.

[0011] Figure 1 Fig. 1 is a schematic diagram of an embodiment of the shipyard steel truss scheme automatic optimization method based on the genetic algorithm in the present application; Figure 2 Fig. 2 is a schematic diagram of an embodiment of the shipyard steel truss scheme automatic optimization system based on the genetic algorithm in the present application; Figure 3 Fig. 3 is a structural schematic diagram of the shipyard steel truss scheme automatic optimization equipment based on the genetic algorithm in the present application. DETAILED DESCRIPTION

[0012] The present application provides a shipyard steel truss scheme automatic optimization method and system based on the genetic algorithm. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0013] For the sake of understanding, the specific process of the embodiments of the present application is described below. Please refer to Figure 1 An embodiment of the shipyard steel truss scheme automatic optimization method based on the genetic algorithm in the present application includes: Step S101, collect the shipyard steel truss structure parameters, establish an engineering experience database, and obtain initial component section parameters; Step S102, parameterize modeling of the steel truss structure based on the initial component section parameters, generate structure attributes and generate various load cases according to the specifications, and obtain a structure calculation model; Step S103, import the structure calculation model into finite element software, perform internal force analysis, and obtain component internal force data; Step S104, perform specification checking for each component according to the component internal force data, calculate various stress ratios, and obtain component control stress ratios; Step S105, set decision variables and constraint conditions based on the component control stress ratios, perform multi-objective optimization iteration through the NSGA-II algorithm, and obtain a Pareto optimal solution set; Step S106, perform three-dimensional modeling and document generation on the Pareto optimal solution set, and obtain final optimization design results.

[0014] It can be understood that the execution subject of the present application can be a shipyard steel rack scheme automatic optimization system based on a genetic algorithm, and can also be a terminal or a server, which is not limited here. The server is taken as an example for description in the embodiments of the present application.

[0015] Specifically, by establishing a structured engineering parameter collection system, the system will collect key data including steel rack calculation span, column height, truss height, load information (such as roof dead load, live load, wind load parameter, earthquake load parameter), crane parameter (crane lifting capacity, crane track spacing, crane wheel spacing, etc.), and civil engineering parameter (such as foundation form, foundation bearing capacity, etc.). These data are input into the system through a unified data interface and are subjected to validity test by a data preprocessing module. In this way, it can be ensured that the collected data are complete and accurate, and at the same time, errors caused by manual operation can be avoided. Through these high-quality engineering data, the system can further establish a multi-dimensional engineering experience database, which is stored in categories according to different workshop types, height intervals, span intervals, crane lifting capacities, etc. The database not only includes geometric parameters of component sections, material parameters, component connection modes, etc. of successful cases, but also combines the experience judgment and matters needing attention of design experts. This enables the system to quickly extract similar schemes from historical cases as initial design parameters when facing new design requirements, thereby significantly improving the efficiency and quality of subsequent optimization. Parametric modeling of the steel rack structure based on initial component section parameters is started. In this stage, the system uses the preliminary collected parameters and historical design data in the database to quickly generate a calculation model of the steel rack structure by using the parameterized modeling technology and the Rhino / Grasshopper software platform. The model includes information such as the geometric topology structure of the steel rack, the component section attribute, the material attribute, the boundary condition and the internal constraint. The advantage of parameterized modeling is that it can flexibly control the geometric size and section attribute of each component, and the system automatically generates all necessary component node connection relationships to ensure the integrity of the model and the accuracy of the calculation. After completing the parameterized modeling, the system automatically generates multiple load cases according to relevant design specifications such as the Steel Structure Design Standard (GB50017-2017), including dead load, live load, wind load, crane load, earthquake load, etc. For live load and crane load, the system will generate multiple load case schemes by calculating the unfavorable arrangement to ensure that the most unfavorable internal force effect can be obtained. Through the establishment of the model in this stage, a solid foundation is laid for subsequent internal force calculation and optimization analysis.

[0016] The structural calculation model is imported into the finite element analysis software ANSYS for internal force analysis. Through seamless docking with ANSYS, the system automatically converts the geometric model and load information into APDL language required by ANSYS, handles local coordinate system definition, component freedom setting, material properties, etc., and ensures the accuracy and reliability of the calculation model. Through the solver of ANSYS, the system analyzes each independent load case separately, calculates the displacement, internal force, stress and other response indicators of each component under different load conditions. For load conditions with multiple unfavorable arrangement schemes, the system will automatically perform multiple calculations and extract the maximum internal force value of each component under each arrangement, so that the internal force envelope value of each component can be obtained, providing complete data support for subsequent checking.

[0017] According to the relevant design specifications such as "Steel Structure Design Standard" (GB50017-2017), the standard checking of various components is carried out, the different stress ratios are calculated, and the control stress ratio of the component is obtained. Through the load combination rule library, the system can automatically generate different load combinations to calculate the stress of the component under normal use limit state and bearing capacity limit state. The system will calculate the axial compression stability, axial tension strength, bending strength, shear strength, etc. of each component, and finally obtain the stress ratio of each component, and extract the maximum value as the control stress ratio of the component. This process is a crucial link in structural design, which ensures that each component can meet the design requirements under different working conditions.

[0018] Based on the control stress ratio of the component, the system sets decision variables and constraint conditions, and uses NSGA-II algorithm for multi-objective optimization iteration. The decision variables include the cross-section parameters of the component, and the stress ratio is used as the constraint condition. The optimization target is the total weight of the steel structure and the surface area of the component (especially related to the spraying cost). Through genetic algorithm, the system generates multiple different component cross-section schemes, performs a series of calculations such as parameterized modeling, finite element analysis and standard checking, and obtains the evaluation index of each scheme. Through crossover and mutation operations of genetic algorithm, the optimization algorithm can generate a series of schemes with good performance, and finally extract the optimal solution set that meets the constraint condition, i.e. the Pareto optimal solution set. These solution sets achieve the best balance between steel consumption and surface area, providing multiple optimization schemes for designers to choose from.

[0019] According to the obtained Pareto optimal solution set, three-dimensional modeling and document generation are performed. Through the Rhino software environment, the system generates a three-dimensional entity model of the final optimized design results, and color labels each component to show different stress ratio intervals. In this way, designers can intuitively understand the stress distribution and potential optimization space of the structure. The system also provides various charts, such as component stress ratio distribution histogram, component weight proportion pie chart, and cross-section usage frequency chart, to help designers analyze the optimization results comprehensively. Finally, the system automatically generates design documents such as calculation sheets, drawings, and lists, supports PDF and CAD format export, and ensures that the design scheme can be directly applied to actual engineering and smoothly enter the bidding and construction stages.

[0020] In the embodiments of the present application, In a specific embodiment, the process of step S101 can specifically include the following steps: Data collection is performed on the calculation span, column height, truss height, load information and crane parameters of the shipyard steel rack to obtain original structure parameters; Data preprocessing and effectiveness test are performed on the original structure parameters to obtain effective structure parameters; A multi-dimensional structured database is constructed based on the effective structure parameters, and the classified structure data is stored according to the workshop type, height interval, span interval and crane lifting capacity to obtain the classified structure data; The classified structure data is associated and mapped with the component cross-section geometric parameters, material parameters and connection mode information to obtain a parameter relationship mapping table; A parameter retrieval algorithm is established according to the parameter relationship mapping table to calculate the similarity between the input parameters and the historical design scheme, and obtain the most similar historical case; The cross-sectional size and plate thickness parameters of the column and beam components are extracted from the most similar historical case to obtain the initial component cross-sectional parameters.

[0021] Specifically, the original structural parameters are obtained through systematic data collection of the shipyard steel bent frame's calculated span, column height, truss height, load information, and crane parameters. This data collection process is the foundation of the entire optimization design process. The accuracy and comprehensiveness of this process ensure the reliability of subsequent optimization calculations. The shipyard steel bent frame parameter collection system must include multiple aspects of information, including the calculated span of the steel bent frame, which affects the overall structural span size; column height and truss height, which directly affect the structural stability and load-bearing capacity; load information, including roof dead load, live load, wind load, and seismic load, which are crucial to the structural stress state; and crane parameters, including crane lifting capacity, track gauge, and wheelbase, which influence the strength and load-bearing capacity of components during steel bent frame design. Crane parameters are particularly important because the distribution of crane loads affects the internal force distribution and stability of various structural components. Through a unified data interface, these original parameters are input into the system and transmitted to the data preprocessing module for preliminary verification to ensure their validity and integrity. The purpose of data preprocessing is to verify the validity of collected parameters and eliminate invalid or erroneous data to ensure the accuracy of subsequent calculations. For example, during the load information collection process, the system verifies the rationality of each parameter value, eliminating extreme values ​​or values ​​that do not conform to actual working conditions. This ensures that the resulting effective structural parameters accurately reflect the structural design requirements in actual applications. After completing data preprocessing, the system constructs a multidimensional structured database based on the effective structural parameters, using a relational database structure to store the data. The database's structure follows specific classification criteria, storing key parameters such as workshop type, height range, span range, and crane capacity to ensure efficient data retrieval for different design scenarios. For example, the database stores commonly used structural parameters for different workshop types. Similarly, structures with different heights and spans are categorized and stored accordingly. This multidimensional, categorized storage not only improves data access efficiency but also facilitates subsequent parameter matching. Each classification contains not only basic cross-sectional geometry parameters, such as H-beam height, web thickness, flange width, and flange thickness, but also material parameters (such as steel strength grade, elastic modulus, and Poisson's ratio), component connection methods (such as rigid and hinged connections), and detailed node drawings. This stored data encompasses all critical parameters potentially used in the design process, ensuring a diverse and rich selection of design options.

[0022] On the basis of the database, the system associates and maps the classification structure data with the geometric parameters of the component cross-section, material parameters, and connection methods through a parameter relationship mapping table, thereby establishing a comprehensive parameter relationship network. The establishment of this mapping relationship enables the system to quickly retrieve similar cases in historical design schemes when a new set of design parameters is input, thereby providing reasonable initial parameters for the design. The core of this process is to form a clear parameter correspondence relationship through the mapping table to ensure effective data matching and conversion between different design schemes. For example, when a specific span and column height are input, the system can find the most similar historical design according to the mapping relationship in the database and extract the corresponding component cross-section parameters and material information. To further improve the efficiency of data retrieval, the system designs a parameter retrieval algorithm based on the parameter relationship mapping table. The purpose of this algorithm is to find the historical case that best meets the current design requirements by calculating the similarity between the input parameters and the historical design schemes. Specifically, when the designer inputs a set of preliminary parameters, the system calculates the similarity between this set of parameters and all historical cases in the database, usually using distance measurement-based methods such as Euclidean distance or weighted similarity to measure the similarity between the input parameters and the historical design schemes. The calculation result shows that the system returns a most similar historical case, which usually provides a reasonable initial scheme for the current design. In this way, the designer does not need to start from scratch but can draw on existing successful cases, saving a lot of design time and effort.

[0023] From the most similar historical case, the system extracts parameters such as the cross-sectional size of the column and beam components and the thickness of the plate, which are used as the initial design parameters of the components. These initial parameters play an important role in the subsequent parametric modeling and optimization process. By drawing on the design schemes that have been verified in historical cases, the selection of initial parameters is more reasonable and efficient. For example, if the H-shaped steel component used in a historical case under a certain span and column height performs well in terms of bending and shear resistance, these historical parameters will provide a reliable starting point for the current design, avoiding the need to perform extensive derivation and verification from scratch.

[0024] In a specific embodiment, the process of performing step S102 can specifically include the following steps: Import the initial component cross-section parameters into the Rhino / Grasshopper platform, build the geometric element parameter equations of the column, beam, and support, and obtain the parameter control equation set; Generate the geometric topology and node connection relationship of the steel bent frame structure based on the parameter control equation set, and obtain the structure geometric model; Assign component cross-section properties, material properties, boundary conditions, and internal constraints to the structure geometric model, and obtain the complete structure model; Generate constant load working condition, wind load working condition and seismic load working condition, get foundation load working condition; Calculate the unfavorable arrangement of live load in the foundation load working condition, and calculate the multi-position wheel pressure of crane load to get the extended load working condition; Convert the extended load working condition into a standardized data format and integrate it with the complete structure model to get the structure calculation model.

[0025] Specifically, the initial member section parameters are imported into the Rhino / Grasshopper platform, and parameterized modeling is performed using the platform. First, based on the initial member section parameters obtained in the previous step, the system imports these parameters into the Rhino / Grasshopper platform to use the powerful modeling functions of the platform for structure modeling. The advantage of Rhino / Grasshopper is that it can generate geometric elements and control equations related to the input parameters, which can accurately define the geometric form and properties of steel batten frame members such as columns, beams, and supports. In this stage, the system constructs parameter equations for geometric elements such as columns, beams, and supports by inputting initial member section parameters, ensuring that the geometric form of the structure can be flexibly adjusted according to these parameters. For example, by inputting parameters such as the height, web thickness, and flange thickness of an H-shaped steel, Rhino / Grasshopper can generate the corresponding geometric model based on these data and form a set of parameter control equations that determine the geometric size and structural characteristics of the member under different conditions. Based on the established parameter control equation set, the system further generates the geometric topology of the steel batten frame structure and determines the node connection relationship. Geometric topology refers to the connection method between steel batten frame members and their relative position in space, while the node connection relationship defines the connection method between different members, such as rigid connection or hinged connection. In this process, the system automatically determines the connection method and connection nodes between members based on their geometric parameters, forming a complete steel batten frame structure geometric model. For example, when generating the connection between the column and the beam, the system will automatically set the connection nodes and connection types of the column and beam based on the input member parameters to ensure the continuity of the model and the integrity of the structure. The key to this stage is to ensure that the geometric relationship and node connection between all members can be adjusted according to the initial input parameters through parameterized modeling, and all changes remain consistent.

[0026] After constructing the geometric topology and node connection relationships of the structure, the system enters the stage of assigning member section properties, material properties, boundary conditions, and internal constraints, completing the generation of the complete structure model. The purpose of this step is to give each member the required physical properties, enabling the model to have the conditions for structural calculation. For example, for each column, beam, support, and other members, the system will assign the corresponding section properties (such as section area, moment of inertia, section modulus, etc.), material properties (such as steel strength grade, elastic modulus, etc.), and boundary conditions (such as support method, constraint conditions, etc.) according to the initial parameters input previously. In addition, internal constraints such as member stability constraints, interactions between members and support points, etc. will also be defined in this stage to ensure that the structure model has reasonable mechanical behavior during analysis. For example, the system may set fixed boundary conditions for each column, while setting different support conditions for beams, which will affect the subsequent mechanical analysis and calculation results. Through this process, the final structure model not only contains geometric shapes, but also has necessary physical properties and boundary conditions, enabling accurate internal force calculation in subsequent finite element analysis. After completing the physical property assignment of the structure model, the system generates constant load cases, wind load cases, and seismic load cases according to relevant design standards and summarizes them as basic load cases. These load cases are indispensable in structural design, representing the influence of external loads on the structure under different conditions. The constant load case considers the self-weight of the steel truss structure and other long-term existing loads; the wind load case considers the load generated on the structure due to wind force; and the seismic load case considers the force on the structure under earthquake action. In this stage, the system will automatically generate these load cases according to the requirements of the "Steel Structure Design Standard" (GB50017-2017), "Building Seismic Design Specification" (GB50011-2010), etc., and ensure that these load cases meet the design requirements of the structure.

[0027] After obtaining the basic load cases, the system further calculates the unfavorable arrangement of live loads in the basic load cases and the multi-position wheel pressure calculation of crane loads to obtain extended load cases. The unfavorable arrangement of live loads refers to selecting the most unfavorable load combination among multiple possible load arrangements to ensure the stability of the structure under the most unfavorable working condition. The multi-position wheel pressure calculation of crane loads considers the influence of the crane on the structure in different positions, which is particularly important for steel truss structures in shipyards, as the crane load may be concentrated on certain key parts, thereby affecting the load-carrying capacity and stability of the structure. In this stage, the system will automatically calculate the maximum wheel pressure generated by the crane load in different positions and ensure that these calculation results accurately reflect the load effects that may be caused during crane operation.

[0028] The system converts the extended load conditions into a standardized data format and integrates it with the complete structural model to obtain the final structural calculation model. The conversion to the standardized data format is intended to ensure that different load conditions can be processed uniformly, facilitating subsequent finite element analysis and calculations. By integrating the load conditions with the structural model, the system ultimately generates a complete structural calculation model that includes the geometry, physical properties, boundary conditions, load conditions, and internal and external constraints of the steel bent structure, enabling further mechanical analysis in finite element software. This process ensures that the structural calculation model is both accurate and complete, laying a solid foundation for subsequent analysis and optimization.

[0029] In a specific embodiment, the process of executing step S103 may specifically include the following steps: Convert the structural calculation model into APDL language, process the local coordinate system definition, degree of freedom setting and material properties, and obtain the finite element input file; Import the finite element input file into ANSYS software, perform single-condition internal force calculation for each independent load condition, and obtain the single-condition internal force results; Multi-threaded parallel computing technology is used to calculate the internal force results of a single working condition, processing various arrangements of live loads and crane loads to obtain the internal force matrix; Based on the internal force matrix, the internal force envelope value extraction algorithm is executed to calculate the maximum internal force value of each component under all possible working conditions and obtain the internal force envelope value; The internal force envelope values ​​are stored according to component numbers and load conditions, and an internal force data index system is established to obtain an internal force database; The axial force, shear force, bending moment internal force components and key node displacement values ​​of each component are extracted from the internal force database to obtain the component internal force data.

[0030] Specifically, the APDL language conversion of the structural calculation model, the execution of the finite element analysis, and the processing of the internal force calculation are the key steps to realize the structural optimization. First, in the process of APDL language conversion of the structural calculation model, the system imports the previously generated structural calculation model into the framework of finite element analysis. The APDL (ANSYS Parametric Design Language) language is the input language of the ANSYS finite element software, which is used to define the geometric shape, material properties, boundary conditions, load cases, etc. of the model. In this stage, the system converts the geometric model of the structure and various properties (such as the cross-sectional size of the member, material properties, etc.) into APDL code that meets the requirements of the ANSYS solver, ensuring that the structural model can be correctly parsed and calculated in the ANSYS software. In particular, when performing APDL language conversion, the system not only needs to handle the definition of the local coordinate system, but also needs to set the degrees of freedom of each member, ensuring that the model can correctly represent the physical properties and constraints of the structure during the analysis process. The definition of the local coordinate system is very important, as it ensures that the force calculation of each member in the local coordinate system is consistent with the global coordinate system; the setting of the degrees of freedom determines the deformation mode (e.g., translation, rotation, etc.) of each node in the model; the setting of the material properties involves the elastic modulus, Poisson's ratio, strength, etc. of the steel, which are indispensable parameters in finite element analysis. Through these conversions, the system finally generates a finite element input file that can be directly imported into the ANSYS solver.

[0031] After importing the finite element input file into the ANSYS software, the system performs single-condition internal force calculation for each independent load case. Each load case represents the stress state of the structure under specific external loads, such as dead load, wind load, live load, crane load, etc. For each load case, ANSYS software calculates the displacement, stress, strain and other responses of the structure under that condition by solving the finite element equation. This process is crucial for calculating how the structure responds to external loads under various conditions. The system analyzes each condition separately and obtains internal force results, such as the axial force, shear force, bending moment and other data of each component in the structure. Through this process, the system can obtain the specific stress conditions of the structure under each load, forming a multi-dimensional internal force data set. To improve computational efficiency, the system uses multi-thread parallel computing technology, especially when dealing with multiple arrangements of live loads and crane loads. Since there may be multiple different arrangements of live loads and crane loads, such as the crane moving to different positions, which will affect the stress of each component, it is necessary to calculate the internal force under each arrangement. Using parallel computing technology, the system can simultaneously process multiple load arrangement schemes, greatly improving computational efficiency. For example, when analyzing the impact of crane load on the structure, the system generates multiple load schemes for different crane positions and simultaneously calculates these schemes to quickly obtain the internal force matrix under each load arrangement. These internal force matrices contain the internal force responses of each component under different arrangements, providing comprehensive data support for subsequent internal force analysis.

[0032] After obtaining the internal force matrix under each load case, the system continues to perform the internal force envelope value extraction algorithm based on the internal force matrix. The extraction of the internal force envelope value is an important step in the analysis, because under complex load combinations, the member may have different maximum internal forces under different load cases, and the internal force envelope value is the maximum internal force value extracted from all possible cases. Through the extraction of the internal force envelope value, the system can obtain the maximum internal force value of each member under all load cases, which represents the stress condition of the member under the most unfavorable condition and is a crucial parameter in the design. For example, under certain load cases, some members may be subjected to greater axial force or bending moment, while other load cases may produce smaller internal forces. Through the internal force envelope value, the system can ensure that the design takes into account all possible stress conditions, thereby ensuring the stability of the structure under the most unfavorable condition. The system stores the internal force envelope value by member number and load case, establishing an internal force data index system. This internal force database is the core data support for the entire design optimization process, as it not only contains the maximum internal force value of each member under each load case, but also allows efficient retrieval by member number. Through this classified storage method, the system can quickly find the internal force response of each member under different load cases and perform further analysis and optimization as needed. The establishment of the internal force database makes the subsequent optimization process more efficient and accurate, providing a solid foundation for the specification checking and optimization design of members.

[0033] The system extracts the axial force, shear force, bending moment internal force components of each member and the key node displacement value from the internal force database to obtain the complete internal force data of the member. Axial force, shear force and bending moment are the most commonly used internal force indicators in structural design, which can accurately reflect the stress of each member under different loads. Key node displacement values can provide information on structural deformation, which is very important for evaluating the stability and service performance of the structure. Through these internal force data, designers can evaluate the safety and stability of each member and provide a basis for subsequent optimization iterations. These data will also be used in specification checking and multi-objective optimization to ensure that the structural design meets all constraint conditions while achieving optimal performance.

[0034] In a specific embodiment, the process of performing step S104 can specifically include the following steps: Based on the internal force data of the member, load combinations under normal use limit state and bearing capacity limit state are performed based on the built-in load combination rule library to obtain combined internal forces; Based on the combined internal forces and the geometric parameters of the member cross section, the cross section characteristics including cross section area, moment of inertia, cross section modulus are calculated to obtain cross section characteristic data; According to the cross section characteristic data, the length of the member, the support condition and the connection mode, the slenderness ratio and the length coefficient of the member are calculated to obtain the stability parameters of the member; The axial pressure stability check, axial tension strength check, bending strength check and shear strength check are performed on the combined internal force to obtain the partial stress ratio; Based on the partial stress ratio, the combined stress ratio calculation under complex stress state is performed to obtain the combined stress ratio of each component. The maximum value of each component is extracted from the partial stress ratio and the combined stress ratio, and a stress ratio data structure is established to obtain the component control stress ratio.

[0035] Specifically, the system will combine the loads of the structure according to the built-in load combination rule library to obtain the combined internal force. When performing load combination, the system will consider the load combination of normal use limit state and bearing capacity limit state according to the requirements in the structure design specification. Under the normal use limit state, the load combination mainly focuses on the load conditions that the structure may encounter in daily use, including dead load, live load, wind load, etc. The combination of these loads is mainly to ensure the stability and service performance of the structure in normal use. Under the bearing capacity limit state, the strength and stability of the structure under extreme load are mainly considered, for example, considering the most unfavorable working condition, such as the combination of crane load and seismic load. Through the load combination rule library, the system can automatically generate the combined internal force under different load working conditions and calculate the internal force response of the structure under these combined loads. In this way, the system not only ensures the safety of the structure under normal working conditions, but also can deal with the risks that may occur under extreme working conditions, ensuring the comprehensiveness and safety of the design. Based on the obtained combined internal force data, the system continues to calculate the cross-sectional characteristics according to the cross-sectional geometric parameters of the components, including cross-sectional area, moment of inertia and cross-sectional modulus, and then obtains the cross-sectional characteristic data. Cross-sectional characteristics are important indicators for evaluating the bending, shear and compression resistance of components in structural design. Through the analysis of the geometric data of the component cross section, the system can calculate the cross-sectional area, moment of inertia and cross-sectional modulus of each component, which are key parameters. These data are crucial for subsequent stress analysis and checking. For example, the moment of inertia reflects the bending resistance of the component, the cross-sectional modulus represents the bending strength of the component, and the cross-sectional area is directly related to the carrying capacity of the component. These cross-sectional characteristic data provide a basis for subsequent specification checking and are an important basis for ensuring the stability and strength of the structure.

[0036] According to the calculated cross-section characteristic data, the system further calculates the stability parameters of the members, including the slenderness ratio and the length coefficient of the members. The slenderness ratio is an important indicator to measure the stability of the members, which represents the ratio of the length of the member to the minimum size of its cross-section, and is usually used to evaluate the buckling stability of members such as columns and beams. The larger the slenderness ratio, the worse the stability of the member, and it is prone to buckling failure. The length coefficient is determined according to the support conditions and connection methods, considering the actual performance of the member under different support conditions. For example, when the member is fixed at both ends, its length coefficient is small, indicating that its stability is strong; while when the member is hinged at both ends, its length coefficient is large, indicating that its stability is poor. By calculating the slenderness ratio and the length coefficient, the system can judge the stability of each member under different load cases, and provide key parameters for subsequent checking.

[0037] After obtaining these stability parameters, the system begins to perform standard checking, and performs axial compression stability checking, axial tension strength checking, bending strength checking and shear strength checking on the combined internal force, to obtain the partial stress ratio of each member. The axial compression stability checking mainly checks whether the member is prone to buckling when subjected to axial compression, and the axial tension strength checking focuses on whether the strength of the member is sufficient when subjected to tension; the bending strength checking evaluates whether the member can withstand the bending moment generated when subjected to bending; the shear strength checking checks whether the member will be sheared when subjected to shear force. Each member has a corresponding stress ratio under different stress conditions, representing the ratio of the actual stress to the allowable stress. If the stress ratio is greater than 1, it indicates that the member is overloaded and needs to be further optimized; if the stress ratio is less than 1, it means that the member is safe under this working condition. When performing these checks, the system will automatically calculate according to the checking formula in the standard specification, and evaluate the stress components such as axial force, bending moment and shear force of each member to obtain the partial stress ratio. After obtaining the partial stress ratio of each member, the system continues to calculate the combined stress ratio, which is used to handle the stress combination under complex stress conditions. Since the member may simultaneously bear axial force, bending moment and shear force under different load cases, a single partial stress ratio cannot fully reflect the comprehensive stress state of the member. Therefore, the system needs to calculate a comprehensive combined stress ratio based on the partial stress ratio by synthesizing the stress ratios of different stress conditions. The combined stress ratio is a key indicator to evaluate whether the member can remain stable and safe under complex stress conditions. For example, under certain load combinations, the member may be subjected to compression and bending at the same time, and the combined stress ratio is needed to consider all stresses comprehensively to ensure that the design of the member meets all stress requirements.

[0038] The system extracts the maximum value of each component from all the sub-stress ratios and combined stress ratios, and stores these maximum stress ratios in a stress ratio data structure, obtaining the control stress ratio of the component. The control stress ratio is a key indicator in structural design, which determines whether the component can safely operate under all possible working conditions. If the control stress ratio exceeds the limit value required by the specification, the component needs to be optimized in design, adjusting the cross-sectional size or material strength to ensure its safety requirements. By extracting the maximum stress ratio of each component, the system can accurately identify the components that may have problems and provide a clear direction for subsequent optimization. Finally, these stress ratio data will help designers evaluate the safety and performance of the overall structure and make adjustments if necessary to ensure the optimality and safety of the design scheme.

[0039] In a specific embodiment, the process of performing step S105 can specifically include the following steps: Set the component cross-section parameters as decision variables, the component stress ratio as a constraint condition, and the total weight of the steel structure and the surface area of the component as optimization objectives to obtain a mathematical model of the optimization problem; Based on the mathematical model of the optimization problem and the component cross-section parameter design gene coding scheme in the experience database, an initial population coding is obtained; Perform crossover and mutation genetic operations on the initial population coding, and handle the constraint violation through the penalty function method to obtain a new generation population; Repeat the parameterized modeling, finite element analysis and specification checking process for each individual in the new generation population to obtain the evaluation index of each individual; Based on the evaluation index, perform population screening using non-dominated sorting and crowding degree calculation, and implement adaptive crossover rate and mutation rate strategies to obtain an evolved population; Convergence monitoring is performed on the evolved population, and when the optimal solution of consecutive generations does not improve significantly, the iteration is stopped, the steel consumption and surface area optimal balance scheme that satisfies the stress ratio constraint is extracted from the last generation population, and a Pareto optimal solution set is obtained.

[0040] Specifically, by setting the member cross-section parameters as decision variables, the member stress ratios as constraint conditions, and the total weight of the steel structure and the member surface area as optimization objectives, a mathematical model of the optimization problem is constructed. The goal of this model is to select the cross-section parameters of the members reasonably, so that the overall weight and the member surface area of the steel structure are optimized, while ensuring that the stress ratio of each member meets the requirements of the design specification, guaranteeing the safety and stability of the structure. The decision variables refer to the parameters that can be adjusted during the optimization process, which in the case of a steel bent structure, mainly refer to the cross-sectional dimensions of the members and the selection of materials; while the constraint conditions refer to the design requirements that must be met, such as the maximum stress ratio of the members not exceeding the limit value specified in the specification. In the mathematical model of the optimization problem, the total weight of the steel structure and the member surface area are used as optimization objectives, with the goal of reducing material usage and thus reducing costs, while ensuring the safety of the structure. Based on the mathematical model of the optimization problem and the member cross-section parameters in the experience database, a gene coding scheme is designed, and an initial population code is generated accordingly. The gene coding scheme converts the cross-section parameters of each member (such as the height of H-shaped steel, web thickness, flange width, etc.) into binary or real number coding, so that each individual in the genetic algorithm is represented as a gene group composed of multiple cross-section parameters. The initial population is initialized by randomly selecting a certain number of individuals and based on the gene coding of these individuals. These individuals represent different design schemes, i.e., different combinations of member cross-section parameters, which will be evaluated and improved through the genetic algorithm during the optimization process. After the initial population code is generated, the system generates a new population by performing genetic operations such as crossover and mutation. The crossover operation simulates gene recombination in natural selection, exchanging parts of the genes of two individuals to produce new design schemes; while the mutation operation randomly changes certain parts of the individual's genes to produce diversity and prevent falling into local optimal solutions. To handle individuals that violate the constraint conditions, the system uses a penalty function method. Specifically, if the stress ratio of a member of an individual exceeds the specification limit, the fitness value of that individual will be penalized, with the degree of punishment being proportional to the degree of exceeding the specification, thus ensuring that the safety and feasibility of the design are always considered during the genetic operation process. Through crossover, mutation, and the penalty function method, the system can generate a new generation of population, which contains multiple improved design schemes.

[0041] In each generation of the genetic algorithm, the system performs parameterized modeling, finite element analysis, and code checking for each individual to calculate its evaluation index. Specifically, the system constructs a corresponding steel bent structure model based on the current individual's gene coding, performs finite element analysis to calculate the internal force, stress, and other response results of the structure under different load cases. Then, the system checks the components according to the code requirements to obtain stress ratios, stability, strength, and other indicators, and evaluates the individual's performance based on these indicators. The evaluation index includes not only the stress ratio of the components, but also the optimization objectives - the total weight of the steel structure and the surface area of the components, which are the parameters to be minimized in the optimization process. Through the evaluation of each individual, the system can identify which schemes can effectively reduce material usage and structure weight while meeting design requirements.

[0042] After obtaining the evaluation index of each individual, the system uses non-dominated sorting and crowding distance calculation to screen the population. Non-dominated sorting is used to sort individuals in the population according to their performance, giving priority to those individuals that perform well on multiple objectives; while crowding distance calculation is used to assess the distribution of individuals in the objective space, ensuring the diversity of the population, avoiding the concentration of all individuals near a certain target value, thereby increasing the range and depth of search. Through these two ways, the system can retain the best individuals and eliminate the poor ones, thereby improving the overall quality of the population. At the same time, the system dynamically adjusts the crossover rate and mutation rate according to the stage of evolution to achieve a balance between local search and global search. For example, in the early stages of evolution, higher crossover rate and mutation rate help to generate diverse design schemes and explore a wider design space; while in the later stages, lower crossover rate and mutation rate help to refine the search and refine the best solution.

[0043] During the iterative process of the genetic algorithm, the system monitors the convergence of the evolving population to determine whether it has entered a local optimal solution. When the optimal solution of consecutive generations does not improve significantly, the system stops iteration and extracts the optimal solution from the last generation of population. The optimal solution set is composed of design schemes that find the best balance between optimization objectives (such as steel usage, component surface area) and constraint conditions (such as component stress ratio). These schemes represent different optimization designs of steel bent structures, providing designers with multiple scheme options and minimizing costs while ensuring structural safety.

[0044] From these optimal solutions, the system can extract the optimal balanced solution that meets all stress ratio constraints, i.e. the Pareto optimal solution set. Each solution in the Pareto optimal solution set achieves optimality on different optimization objectives, and no other solution can improve a certain objective without sacrificing other objectives. These Pareto optimal solutions provide designers with multiple alternative schemes, allowing them to choose the most suitable design scheme according to actual requirements.

[0045] In a specific embodiment, the process of performing step S106 can specifically include the following steps: The Pareto optimal solution set is converted into three-dimensional entity model data, geometric construction is performed in the Rhino software environment, and a three-dimensional model of the steel bent structure is obtained; The members in the three-dimensional model of the steel bent structure are color-coded according to the stress ratio interval, and member numbers and cross-section information are added to obtain a visual structure model; Based on the visual structure model, a member stress ratio distribution histogram, a member weight proportion pie chart, and a cross-section usage frequency statistical chart are generated to obtain a data analysis atlas; Different optimization schemes in the Pareto optimal solution set are compared and analyzed through a radar chart for multi-index comparison, which assists in scheme selection, and a scheme comparison report is obtained; According to the optimal scheme in the scheme comparison report, a calculation book is automatically generated, which records calculation parameters, load conditions, calculation process, and checking results in detail to obtain a design calculation document; Based on the design calculation document, plan view, elevation view, section view, node detail view, material usage statistical table, and member detail table that meet national standards and specifications are generated to obtain the final optimized design results.

[0046] Specifically, the Pareto optimal solution set represents multiple optimized design schemes that achieve the best balance between the total weight of the steel structure, the surface area of the members, and other targets, while meeting all stress ratio constraints. The relevant parameters of each design scheme are extracted from the Pareto optimal solution set, including the cross-sectional size of each member, material selection, etc., and these parameters are converted into three-dimensional entity model data recognizable by Rhino software. In the Rhino environment, through parameterized modeling technology, the system automatically builds a three-dimensional geometric model of the steel bent structure using the imported parameter data. The powerful functions of Rhino can generate a real and interactive structure model according to the geometric size of the members and the node connection relationship, which facilitates visual analysis and further optimization of the steel bent structure by designers.

[0047] After obtaining the three-dimensional model of the steel truss structure, the system further visualizes the components in the model. By color-coding based on stress ratio intervals, it highlights the stress conditions of each component under different load cases. Through this color coding, designers can intuitively identify which components bear more stress and which components may have potential risks in the design. Each component is also labeled with a number and cross-section information, which is crucial for subsequent design optimization and construction drawing generation, ensuring that the design parameters of each component are accurately transmitted. Through color coding and information labeling, the final visualized structure model not only provides a complete three-dimensional structure view for designers, but also assists them in evaluating structural performance more efficiently. Based on the visualized structure model, the system can also generate a series of data analysis charts to assist designers in further analyzing the performance of the structure. Specifically, the system generates component stress ratio distribution histograms based on the stress ratio of each component, providing a visual display of the stress conditions of each component in the structure. These histograms help designers evaluate the stress distribution of different components under different load cases, and then determine which components need further optimization. At the same time, the system generates a component weight proportion pie chart to show the material proportion of different components in the structure, which is important for evaluating the material utilization and cost control of the steel structure. In addition to the weight proportion chart, cross-section usage frequency statistical charts are also generated, which help designers understand the frequency of different cross-section types in the entire structure, further guiding the optimization of the design scheme. Through these data analysis charts, designers can more clearly understand the key parameters in the optimization results, and then make reasonable design decisions.

[0048] After completing the visual display of the structure model, the system also conducts multi-index comparative analysis of different optimization schemes in the Pareto optimal solution set through radar charts. Radar charts can clearly show the performance of each optimization scheme in multiple indicators (such as steel consumption, component surface area, stress ratio, etc.), helping designers intuitively compare the advantages and disadvantages of different schemes. This multi-dimensional comparison method can effectively assist designers in selecting schemes, thereby providing strong support for the final design decision. For example, designers can observe through the radar chart that in a certain scheme, the steel consumption and surface area are small, but the stress ratio may be slightly insufficient, while another scheme may perform well in stress ratio. In this way, designers can choose the most suitable optimization scheme according to different design requirements.

[0049] After selecting the optimal scheme, the system will automatically generate a calculation book that details the calculation parameters, load conditions, calculation process, and checking results of the scheme. The calculation book is an important part of the design document, as it not only records all the technical details involved in the entire design process, but also provides a basis for subsequent construction and acceptance. In the calculation book, the system will list the internal force calculation results for each load condition in detail, showing the stress and deformation of the components under different load conditions, and checking the design of each component to ensure that it meets the relevant code requirements. The calculation book will also record all the parameters and methods used in the structure optimization process, ensuring the traceability and transparency of the design. Based on the content of the calculation book, the system will automatically generate design drawings that meet the national standard specifications. These drawings include plan views, elevation views, section views, node detail views, and other detailed information that fully demonstrates the geometric shape, component position, and connection method of the steel truss structure. At the same time, the system will generate material consumption statistics and component detail tables to provide a basis for subsequent construction and bidding. These drawings and tables will be generated in accordance with the format of national and industry standards, ensuring that the design document meets the relevant code requirements and can seamlessly interface with actual engineering construction. Finally, all generated documents and drawings will be provided to the design personnel and construction parties as the final optimized design results, ensuring that the optimized design can be successfully implemented.

[0050] The above describes the method for automatically optimizing a shipyard steel truss scheme based on a genetic algorithm in an embodiment of the present application. The following describes the system for automatically optimizing a shipyard steel truss scheme based on a genetic algorithm in an embodiment of the present application. Please refer to Figure 2 An embodiment of the system for automatically optimizing a shipyard steel truss scheme based on a genetic algorithm in the present application includes: A collection module 201 is configured to collect shipyard steel truss structure parameters, establish an engineering experience database, and obtain initial component cross-section parameters. A modeling module 202 is configured to perform parametric modeling of a steel truss structure based on the initial component cross-section parameters, generate structure attributes, and generate various load conditions according to the specifications, to obtain a structure calculation model. An analysis module 203 is configured to import the structure calculation model into finite element software and perform internal force analysis to obtain component internal force data. A checking module 204 is configured to perform code checking on each component based on the component internal force data, calculate various stress ratios, and obtain component control stress ratios. An iteration module 205 is configured to set decision variables and constraint conditions based on the component control stress ratios, perform multi-objective optimization iteration through an NSGA-II algorithm, and obtain a Pareto optimal solution set. A generation module 206 is configured to perform three-dimensional modeling and document generation on the Pareto optimal solution set to obtain final optimized design results.

[0051] Through the cooperation of the above-mentioned various components, through the shipyard steel rack scheme automatic optimization method based on genetic algorithm, the efficient, accurate and automatic of the steel rack structure design is realized. Firstly, the application ensures the rationality of the initial component section parameter through the collection of the shipyard steel rack structure parameter and the establishment of the engineering experience database. This method breaks through the limitation of traditional design relying on experience, can quickly obtain historical design data and provide basis for new design scheme. Through the combination of the initial component section parameter and the parametric modeling of the steel rack structure, the system can accurately generate the geometric model of the structure and automatically generate various load conditions, so that the whole design process is more systematic and efficient. Compared with the traditional method, this parametric modeling not only reduces the error rate of manual operation, but also makes the response of the structure design under different conditions more accurate. The application introduces a multi-objective optimization algorithm-NSGA-II genetic algorithm in the design. The advantage of this algorithm in multi-objective optimization is particularly prominent, which can optimize multiple objectives such as the weight, component surface area and stress ratio of the steel rack structure, reduce the material usage and cost under the premise of ensuring the safety of the structure. The genetic algorithm makes the optimization process more flexible through the adaptive crossover rate and mutation rate strategy, avoiding the problem that the traditional optimization method may fall into a local optimal solution. By using the multi-objective optimization characteristics of the genetic algorithm, the application can balance between different design schemes and help the designer to select the most cost-effective design scheme. At the same time, the algorithm can screen the population through non-dominated sorting and congestion degree calculation, ensuring the efficiency and diversity of the optimization process and ensuring that the final Pareto optimal solution set has the best performance. By introducing an artificial intelligence model to data mine and pattern recognize the historical design cases. Through the comparison and analysis of the historical design data, the system can quickly find the most similar historical case from the database according to the new input parameter, and extract the initial component section parameter, providing a reasonable starting point for subsequent parametric modeling and optimization iteration. This intelligent data matching process greatly improves the design efficiency, avoids the errors and inconsistencies caused by manual judgment, and ensures the scientificity and rationality of the design scheme.

[0052] The above Figure 2 The shipyard steel rack scheme automatic optimization system based on genetic algorithm in the embodiment of the application is described in detail from the perspective of modular functional entities, and the shipyard steel rack scheme automatic optimization equipment based on genetic algorithm in the embodiment of the application is described in detail from the perspective of hardware processing.

[0053] Figure 3The genetic algorithm-based shipyard steel rack scheme automatic optimization device 300 can have great differences due to different configurations or performances, and can include one or more central processing units (CPUs) 310 (for example, one or more processors) and a memory 320, one or more storage media 330 (for example, one or more mass storage device ends) storing application programs 333 or data 332. The memory 320 and the storage medium 330 can be temporary storage or persistent storage. The programs stored in the storage medium 330 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations in the genetic algorithm-based shipyard steel rack scheme automatic optimization device 300. Further, the processor 310 can be configured to communicate with the storage medium 330 and execute a series of instruction operations in the storage medium 330 on the genetic algorithm-based shipyard steel rack scheme automatic optimization device 300 to realize the steps of the genetic algorithm-based shipyard steel rack scheme automatic optimization method described above.

[0054] The genetic algorithm-based shipyard steel rack scheme automatic optimization device 300 can also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that Figure 3 The genetic algorithm-based shipyard steel rack scheme automatic optimization device structure shown does not constitute a limitation on the genetic algorithm-based shipyard steel rack scheme automatic optimization device provided by the present application, and can include more or fewer components than shown, or combine certain components, or different component arrangements.

[0055] The present application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium, and can also be a volatile computer readable storage medium, and the computer readable storage medium has instructions stored therein, which, when executed on a computer, cause the computer to perform the steps of the genetic algorithm-based shipyard steel rack scheme automatic optimization method.

[0056] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0057] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a shipyard steel rack scheme automatic optimization device based on a genetic algorithm (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0058] The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for automatic optimization of shipyard steel bent scheme based on genetic algorithm, characterized in that: The method comprises: Collect the structural parameters of the shipyard steel bent frame, establish an engineering experience database, and obtain the initial component cross-sectional parameters; Performing parametric modeling on the steel bent structure based on the initial component cross-sectional parameters, generating structural properties and generating various load conditions according to specifications to obtain a structural calculation model; Importing the structural calculation model into finite element software to perform internal force analysis and obtain component internal force data; Performing standard calculations on each component based on the component internal force data, calculating various stress ratios, and obtaining the component control stress ratio; Decision variables and constraints are set based on the component control stress ratio, and a multi-objective optimization iteration is performed through the NSGA-II algorithm to obtain the Pareto optimal solution set; The Pareto optimal solution set is subjected to three-dimensional modeling and document generation to obtain the final optimized design result.

2. The method for automatic optimization of shipyard steel bent scheme based on genetic algorithm according to claim 1, characterized in that: The shipyard steel bent structure parameters are collected to establish an engineering experience database and obtain the initial component cross-sectional parameters, including: Collect data on the calculated span, column height, truss height, load information, and crane parameters of the shipyard steel bent frame to obtain the original structural parameters; Performing data preprocessing and validity testing on the original structural parameters to obtain valid structural parameters; A multi-dimensional structured database is constructed based on the effective structural parameters, and classified and stored according to workshop type, height range, span range, and crane lifting capacity to obtain classified structural data; Associating and mapping the classification structure data with component section geometry parameters, material parameters, and connection method information to obtain a parameter relationship mapping table; Establishing a parameter retrieval algorithm based on the parameter relationship mapping table to calculate the similarity between the input parameters and the historical design solutions and obtain the most similar historical case; The cross-sectional dimensions and plate thickness parameters of the column and beam components are extracted from the most similar historical case to obtain the initial component cross-sectional parameters.

3. The method for automatic optimization of shipyard steel bent scheme based on genetic algorithm according to claim 1, characterized in that: The steel bent structure is parametrically modeled based on the initial component cross-sectional parameters, structural properties are generated, and various load conditions are generated according to specifications to obtain a structural calculation model, including: Importing the initial component cross-section parameters into the Rhino / Grasshopper platform, constructing parametric equations of column, beam, and support geometric elements, and obtaining a set of parameter control equations; Generate the geometric topology and node connection relationship of the steel bent structure based on the parameter control equation group to obtain a structural geometric model; Assigning component cross-sectional properties, material properties, boundary conditions, and internal constraints to the structural geometric model to obtain a complete structural model; Generate dead load conditions, wind load conditions, and earthquake load conditions to obtain foundation load conditions; Perform unfavorable arrangement calculation on the live load in the foundation load condition and perform multi-position wheel pressure calculation on the crane load to obtain the extended load condition; The extended load condition is converted into a standardized data format and integrated with the complete structural model to obtain a structural calculation model.

4. The method for automatic optimization of shipyard steel bent scheme based on genetic algorithm according to claim 1, characterized in that: The step of importing the structural calculation model into finite element software to perform internal force analysis and obtain component internal force data includes: Performing APDL language conversion on the structural calculation model, processing local coordinate system definition, degree of freedom setting and material properties, and obtaining a finite element input file; Importing the finite element input file into ANSYS software, performing single-condition internal force calculation on each independent load condition, and obtaining single-condition internal force results; The internal force results of the single working condition are calculated using multi-threaded parallel computing technology to process various arrangements of live loads and crane loads to obtain an internal force matrix; Executing an internal force envelope value extraction algorithm based on the internal force matrix to calculate the maximum internal force value of each component under all possible working conditions to obtain an internal force envelope value; The internal force envelope values ​​are classified and stored according to component numbers and load conditions, an internal force data index system is established, and an internal force database is obtained; The axial force, shear force, bending moment internal force components and key node displacement values ​​of each component are extracted from the internal force database to obtain component internal force data.

5. The method for automatic optimization of shipyard steel bent scheme based on genetic algorithm according to claim 1, characterized in that: The method of performing standard calculations on each component based on the component internal force data, calculating various stress ratios, and obtaining the component control stress ratio includes: Perform load combinations of the serviceability limit state and the ultimate bearing capacity limit state based on the component internal force data and a built-in load combination rule library to obtain combined internal forces; Calculating cross-sectional properties, including cross-sectional area, moment of inertia, and cross-sectional modulus, based on the combined internal forces and component cross-sectional geometric parameters to obtain cross-sectional property data; Calculate the component slenderness ratio and length coefficient based on the cross-sectional characteristic data, component length, support conditions and connection method to obtain component stability parameters; Performing axial pressure stability calculation, axial tension strength calculation, bending strength calculation, and shear strength calculation on the combined internal force to obtain partial stress ratios; Calculating the combined stress ratio under a complex stress state based on the partial stress ratio to obtain the combined stress ratio of each component; The maximum value of each component is extracted from the partial stress ratios and the combined stress ratios, a stress ratio data structure is established, and the component control stress ratio is obtained.

6. The method for automatic optimization of shipyard steel bent scheme based on genetic algorithm according to claim 1, characterized in that: The decision variables and constraints are set based on the component control stress ratio, and a multi-objective optimization iteration is performed through the NSGA-II algorithm to obtain a Pareto optimal solution set, including: The cross-sectional parameters of the components are set as decision variables, the stress ratio of the components is set as the constraint condition, the total weight of the steel structure and the surface area of ​​the components are set as the optimization targets, and the mathematical model of the optimization problem is obtained; Designing a gene coding scheme based on the mathematical model of the optimization problem and the cross-sectional parameters of the components in the empirical database to obtain an initial population code; Performing crossover and mutation genetic operations on the initial population code, and processing constraint violations by a penalty function method to obtain a new generation population; Repeating the parametric modeling, finite element analysis, and standard verification process for each individual in the new generation population to obtain evaluation indicators for each individual; Based on the evaluation index, non-dominated sorting and crowding calculation are used to screen the population, and adaptive crossover rate and mutation rate strategies are implemented to obtain an evolved population; The convergence of the evolving population is monitored, and when there is no obvious improvement in the optimal solution for multiple generations, the iteration is stopped, and the optimal balance solution between steel consumption and surface area that meets the stress ratio constraint is extracted from the last generation population to obtain the Pareto optimal solution set.

7. The method for automatic optimization of shipyard steel bent scheme based on genetic algorithm according to claim 1, characterized in that: The three-dimensional modeling and document generation of the Pareto optimal solution set to obtain the final optimized design results include: Converting the Pareto optimal solution set into three-dimensional solid model data, performing geometric construction in a Rhino software environment, and obtaining a three-dimensional model of the steel bent structure; Color-marking components in the three-dimensional model of the steel bent structure according to stress ratio intervals, and adding component numbers and cross-sectional information annotations to obtain a visual structural model; Generate a component stress ratio distribution histogram, a component weight ratio pie chart, and a section usage frequency statistical chart based on the visual structural model to obtain a data analysis atlas; Perform multi-index comparative analysis on different optimization schemes in the Pareto optimal solution set through radar charts to assist in scheme selection and obtain a scheme comparison report; Automatically generate a calculation report based on the optimal solution in the solution comparison report, recording the calculation parameters, load conditions, calculation process and verification results in detail to obtain a design calculation document; Based on the design calculation documents, plan drawings, elevation drawings, cross-section drawings, node details, material usage statistics, and component lists that comply with national standards are generated to obtain the final optimized design results.

8. A shipyard steel frame automatic optimization system based on genetic algorithm, characterized by: The method for automatically optimizing a shipyard steel bent scheme based on a genetic algorithm according to any one of claims 1 to 7 is used, wherein the automatic optimization system for a shipyard steel bent scheme based on a genetic algorithm comprises: The acquisition module is used to collect the structural parameters of the shipyard steel bent frame, establish an engineering experience database, and obtain the initial component cross-sectional parameters; A modeling module is used to perform parametric modeling on the steel bent structure based on the initial component cross-sectional parameters, generate structural properties and generate various load conditions according to specifications to obtain a structural calculation model; An analysis module is used to import the structural calculation model into finite element software to perform internal force analysis and obtain component internal force data; A verification module is used to perform standard verification on each component based on the component internal force data, calculate various stress ratios, and obtain the component control stress ratio; An iterative module is used to set decision variables and constraints based on the component control stress ratio, perform multi-objective optimization iterations through the NSGA-II algorithm, and obtain a Pareto optimal solution set; The generation module is used to perform three-dimensional modeling and document generation on the Pareto optimal solution set to obtain the final optimized design results.

9. An automatic optimization device for shipyard steel frame scheme based on genetic algorithm, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the method for automatically optimizing a shipyard steel frame scheme based on a genetic algorithm as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the processor is enabled to execute the method for automatically optimizing a shipyard steel bent scheme based on a genetic algorithm according to any one of claims 1 to 7.

Images