A ship structure design system based on parameterized modeling

Abstract

The application discloses a ship structure design system based on parameterized modeling, and belongs to the technical field of ship structure design, comprising an analytical modeling module, a perception packaging module, a critical positioning module, a parameter locking module, a constraint propagation module, a model generation module, a topology verification module and a specification checking module; the application can improve the coordination and adjustment efficiency of ship structure parameter design, reduce the workload of repeated modification and artificial checking, reduce the risk of structure mismatch caused by local parameter adjustment, improve the accuracy and consistency of three-dimensional geometric model reconstruction; meanwhile, the application can quickly adjust parameters for different design targets, improve the optimization, comparison and iteration efficiency of schemes, reduce design errors and rework, and improve the quality of ship structure design.

Description

Technical Field

[0001] This invention relates to the field of ship structural design technology, and in particular to a ship structural design system based on parametric modeling. Background Technology

[0002] With the development of larger and more complex ships and digital design technology, ship structural design involves a large number of structural objects such as hull plates, skeletons, stiffeners, ribs, longitudinal and transverse members. Their dimensional parameters, positional relationships and connection topologies are strongly correlated. Traditional ship structural design usually relies on computer-aided design software to create a three-dimensional model, and designers adjust the structural dimensions, positions and connection relationships item by item according to design requirements.

[0003] While existing ship structural design systems utilize computer-aided design software for 3D modeling, they lack a unified correlation and driving mechanism between various structural parameters. Adjusting a parameter cannot automatically update related components, resulting in insufficient design coordination, cumbersome adjustment processes, and repeated manual verification for each change, leading to a large workload of repetitive modifications. Furthermore, existing ship structural design systems lack the ability to rapidly adjust parameters for multiple design objectives. When different scheme requirements arise, they cannot easily change parameters in batches and automatically regenerate the model, making the scheme optimization, comparison, and iteration process slow and inefficient. Therefore, we propose a ship structural design system based on parametric modeling. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a ship structure design system based on parametric modeling.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A ship structure design system based on parametric modeling includes an analytical construction module, a perception encapsulation module, a critical positioning module, a parameter locking module, a constraint propagation module, a model generation module, a topology verification module, and a specification verification module. The parsing construction module is used to decompose various structural design specification texts into parameter dependency logic trees, and to construct a numerical potential field based on each specification constraint in the parameter dependency logic tree. The perception encapsulation module is used to obtain the initial parameters of the ship structure, calculate the influence factor matrix of each parameter on the global geometric topology of the model, and then encapsulate each parameter of the ship structure model into parameter-constraint pairs based on the influence factor matrix. The critical positioning module locates the critical equipotential surface that meets all requirements based on the numerical potential field and all parameter-constraint pairs, and at the same time calculates the minimum canonical distance vector between the current parameter-constraint pair and the critical equipotential surface. The parameter locking module is used to calculate the optimal solution set of the design variable that meets the specifications based on the minimum specification distance vector when the user selects the target structural design variable. At the same time, the user can select values ​​within the range of the optimal solution set to obtain a set of compliant parameters. The constraint propagation module is used to take the compliance parameter set as the input basis. When the user modifies any structural parameter, it determines the adaptation strategy of the downstream associated features based on the topology path information in the parameter-constraint pair and completes the topology adaptive correction to obtain the complete parameter set. The model generation module generates a three-dimensional geometric model of the ship structure based on a complete set of parameters and according to the topological path information in the parameter-constraint pairs, in accordance with the corresponding hierarchical order. The topology verification module is used to check the correspondence between the topological connection relationship and parameter constraints of the geometric features in the three-dimensional geometric model of the ship structure one by one, and to check whether there are geometric anomalies caused by parameter propagation. The specification verification module is used to extract the parameters of each component from the three-dimensional geometric model of the ship structure, verify whether they meet all specification requirements one by one, and then generate a verification report and deviation item prompts based on the verification results.

[0006] As a further aspect of the present invention, the specific steps by which the perception encapsulation module encapsulates the parameters of the ship structure model into parameter-constraint pairs are as follows: S1.1: Obtain the initial parameters input during the ship structure design phase, and perform deduplication and engineering value range verification on all initial parameters. Eliminate duplicate definitions and abnormal parameters that exceed the conventional engineering boundaries. Then integrate the remaining initial parameters to form a candidate parameter set. Subsequently, perform sensitivity scanning on the candidate parameter set. During the scanning process, select each candidate parameter as a disturbance variable in turn, apply a quantitative numerical disturbance within its preset range allowed by engineering, and then track the global geometric topology response of the model caused by the disturbance throughout the process, while recording the corresponding topology change data. S1.2: Based on the topology change data of each parameter, calculate the comprehensive global topology impact value corresponding to each parameter, and arrange them according to the parameter index and topology feature dimension to construct the influence factor matrix of each parameter on the global geometric topology of the model. At the same time, based on the influence factor matrix, map the comprehensive global topology impact value of each parameter to the model reconstruction failure probability level, and divide all candidate parameters into hard parameters with strong topology correlation and flexible parameters with weak topology correlation according to the preset engineering probability threshold. S1.3: After completing the classification of parameter topology attributes, match the corresponding geometric topology dependency path and associated constraints for each parameter. At the same time, dynamically calculate the asymmetric allowable fluctuation domain corresponding to the parameter by combining the influence factor matrix. Then, encapsulate the parameter value, geometric topology dependency path and asymmetric allowable fluctuation domain into a unified parameter-constraint pair.

[0007] As a further aspect of the present invention, the specific steps of the critical positioning module in calculating the minimum canonical distance vector of the current parameter-constraint pair from the critical equipotential surface are as follows: S2.1: Based on the numerical potential field and all parameter-constraint pairs constructed by the analytical construction module, a multidimensional parameter space coordinate system is constructed with each parameter as an independent dimension, corresponding one-to-one with the parameter-constraint pairs. The initial value of each parameter is mapped to the reference origin in the multidimensional parameter space coordinate system. At the same time, the allowable fluctuation domain carried in each parameter-constraint pair is mapped to the value range in the corresponding dimension. After completing the parameter space construction and boundary mapping, the overall sampling range is constrained within the value range of each parameter. A differentiated sampling strategy is executed according to the topological attributes of each parameter. Then, sampling is traversed by layering points in each dimension to form a set of sampling points. S2.2: Obtain the parameter combination corresponding to each sampling point and substitute it into the numerical potential field to calculate the potential field value of the sampling point. When the potential field value of any sampling point is greater than or equal to the preset standard critical threshold, the sampling point is determined to be a compliant point; otherwise, it is determined to be a non-compliant point. After completing the calculation of all sampling points, all sampling points are divided into a compliant point set and a non-compliant point set to form a discrete boundary point set of compliant and non-compliant regions. Then, based on the discrete boundary point sets of the two types of regions, a spatial interpolation fitting algorithm is used to construct a continuous critical equipotential surface in the parameter space. During the fitting process, the module uses the gradient direction of the numerical potential field as the fitting guide to correct the surface of the hard parameters. S2.3: After the critical equipotential surface is located, the shortest distance projection calculation is performed on the critical equipotential surface, starting from the reference origin corresponding to the current initial parameter combination, to obtain the minimum gauge distance vector from the reference origin to the critical equipotential surface. Then, based on the boundary of the allowable fluctuation domain of the critical equipotential surface and each parameter, the feasible domain range of parameter design that simultaneously satisfies gauge constraints and topological feasibility is delineated. Finally, the critical equipotential surface, the minimum gauge distance vector, and the feasible domain range of parameter design are sent to the parameter locking module.

[0008] As a further aspect of the present invention, the specific steps by which the parameter locking module obtains the compliant parameter group are as follows: S3.1: Receive the critical equipotential surface, the minimum standard distance vector, and the feasible region of parameter design. Simultaneously, retrieve all parameter-constraint pairs and construct a constraint framework for parameter locking operations using the feasible region of parameter design as the value boundary, the critical equipotential surface as the compliance judgment benchmark, and the minimum standard distance vector as the adjustment direction basis. Then, obtain the target structural design variables selected by the user. For each selected design variable, match its dimension and parameter-constraint pairs in the parameter space, and obtain the corresponding topological attributes, allowed fluctuation domain, and constraint information from them to determine that the target structural design variables are within the feasible region of parameter design. S3.2: Obtain the user-selected optimization objective. When the user-selected optimization objective is to meet the preset specification, the critical value range of the design variable is obtained by following the direction of the minimum specification distance vector pointing to the critical equipotential surface and combining it with the feasible region boundary. When the user-selected optimization objective is the preset optimal margin, the optimal margin value range of the design variable is obtained by following the direction of the minimum specification distance vector away from the critical equipotential surface and combining it with the feasible region boundary. The value range obtained is the optimal solution set of the corresponding target structure. Then, after the user selects the parameter value of the design variable within the optimal solution set, the parameter value is mapped to the corresponding coordinate point in the parameter space. S3.3: Based on the geometric topology dependent path and constraint conditions carried in the parameter-constraint pair, the structural parameters that are related to the design variables of the target structure are reverse-driven to perform linkage compensation. During the compensation process, the overall parameter combination is kept within the compliance range. The corresponding compensation amount is calculated in combination with the allowable fluctuation range of each related parameter. After completing the linkage compensation of all related parameters, a complete compliant parameter set is obtained.

[0009] As a further aspect of the present invention, the specific steps by which the constraint propagation module obtains the complete parameter set are as follows: S4.1: Using the compliant parameter set as the input basis for topology correction operation, each parameter-constraint pair as an independent node of the graph, and the geometric topology dependency path carried in the parameter-constraint pair as the connection edge of the graph, a bidirectional constraint graph is constructed. At the same time, the user-initiated structural parameter modification command is received, the target node of the modified parameter in the bidirectional constraint graph is located, and the topology attribute, allowable fluctuation domain and path information of the parameter-constraint pair corresponding to the target node are obtained. It is determined whether the modified parameter value is within the allowable fluctuation domain of the parameter. If the modified parameter value is within the allowable fluctuation domain, the parameter value of the target node is updated. S4.2: Starting from the updated target node, perform a full-path bidirectional traversal along the connecting edges of the bidirectional constraint graph. During the traversal, sequentially retrieve all downstream feature nodes that are directly or indirectly related to the target node along the topological dependency path. Simultaneously collect the topological association strength, parameter allowable fluctuation range, and current value information of each associated node to form a complete downstream association feature set. Then, based on the downstream association feature set and combined with the topological attributes of each feature node, determine the adaptation strategy one by one. If it is a feature node with strong topological association, it is determined to be a forced following adaptation strategy; otherwise, it is determined to be a flexible adaptive strategy. S4.3: For feature nodes using the forced follow-adaptation strategy, the corresponding linkage correction amount is calculated according to the modification amount of the target node and the topology association ratio; for feature nodes using the flexible adaptive strategy, the minimum topology distortion is used as the optimization objective, and the minimum energy offset is solved within the allowable fluctuation domain. At the same time, the corrected value of each node is verified synchronously during the calculation process. Then, the modified value of the target node, the corrected value of all associated nodes, and the remaining parameter values ​​that are not affected by this modification are integrated to form a complete parameter set.

[0010] As a further aspect of the present invention, the specific steps of the model generation module in generating a three-dimensional geometric model of the ship structure according to the corresponding hierarchical order are as follows: S5.1: Based on the topology path information carried in the parameter-constraint pair, establish the mapping relationship between each parameter and the corresponding geometric feature one by one. At the same time, according to the topology hierarchy attribute in the parameter-constraint pair, extract the parameter subset corresponding to the basic structure from the complete parameter set, and generate the geometric entity of the basic structure with the preset global reference coordinate system of the hull as the positioning reference. After the basic structure is generated, fix its reference plane and axis position. Then, on the generated basic structure positioning reference, according to the component attachment relationship recorded in the topology path information, extract the parameter subset corresponding to the secondary component from the complete parameter set, generate the geometric entity of the secondary structure component in sequence, and then assemble the secondary structure component to the corresponding position of the basic structure. S5.2: Based on the detailed subordination relationships in the topology path information, extract the parameter subsets corresponding to the detailed features from the complete parameter set, and generate detailed geometric features sequentially at the specified positions of the corresponding secondary components or basic structures. After all hierarchical geometric features are generated, they are integrated to form a complete three-dimensional geometric model of the ship structure. Then, based on the topology connection requirements in the parameter-constraint pairs, verify the connection status, fitting accuracy and relative position between each component one by one, check for topology breaks or geometric interference, and after completing the topology consistency verification, obtain the final three-dimensional geometric model of the ship structure.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention obtains initial parameters for ship structural design input, performs deduplication and engineering value range verification, eliminates duplicate definitions and abnormal parameters exceeding conventional engineering boundaries, forming a candidate parameter set. Then, a sensitivity scan is performed, sequentially selecting candidate parameters as perturbation variables, applying numerical perturbations within a preset interval, simultaneously tracking the model's global geometric topological response and recording topological change data, calculating the comprehensive global topological influence value of each candidate parameter, constructing an influence factor matrix and mapping it to the model reconstruction failure probability level, and then classifying strong topologically correlated hard parameters and weakly correlated flexible parameters according to a preset engineering probability threshold. Simultaneously, geometric topological dependent paths, correlation constraints, and asymmetric allowable fluctuation domains are matched to form parameter-constraint pairs. Then, a multidimensional parameter space coordinate system is constructed based on the numerical potential field and parameter-constraint pairs, mapping the initial values ​​to the reference origin and the allowable fluctuation domain to the value interval. Subsequently, differentiated sampling is performed based on the parameter topological attributes, calculating the potential field values ​​of the sampling points and dividing them into compliant and non-compliant point sets, fitting continuous critical equipotential surfaces and obtaining the minimum specification distance vector, and determining the parameters in conjunction with the allowable fluctuation domain boundary. The feasible region is designed, and then the critical value range or margin optimal value range is solved according to the target structural design variables and optimization objectives. After selecting parameter values, the associated structural parameters are compensated in a coordinated manner to obtain a compliant parameter set. Then, a bidirectional constraint graph is constructed with parameter-constraint pairs as nodes and geometric topology dependent paths as connecting edges. According to the structural parameter modification instructions, a full path bidirectional traversal is performed, so that nodes with strong topological association adopt a forced following adaptation strategy, and nodes with weak topological association adopt a flexible adaptive strategy. Finally, based on the topological path information, the basic structure, secondary structural components and detailed geometric features are generated sequentially to form a three-dimensional geometric model of the ship structure. The connection status, fitting accuracy, relative position, topological breakage and geometric interference are verified. The final three-dimensional geometric model of the ship structure can improve the coordination and adjustment efficiency of ship structural parameter design, reduce the workload of repeated modifications and manual verification, reduce the risk of structural mismatch caused by local parameter adjustments, and improve the accuracy and consistency of three-dimensional geometric model reconstruction. At the same time, it can quickly adjust parameters for different design objectives, improve the efficiency of scheme optimization, comparison and iteration, reduce design errors and rework, and improve the quality of ship structural design. Attached Figure Description The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Figure 1 This is a system block diagram of a ship structure design system based on parametric modeling proposed in this invention. Detailed Implementation

[0012] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0013] Example 1, referring to Figure 1 A ship structure design system based on parametric modeling includes an analytical construction module, a perception encapsulation module, a critical positioning module, a parameter locking module, a constraint propagation module, a model generation module, a topology verification module, and a specification verification module. The parsing module is used to decompose various structural design specification texts into parameter dependency logic trees, and to construct a numerical potential field based on each specification constraint in the parameter dependency logic tree.

[0014] The perception encapsulation module is used to obtain the initial parameters of the ship structure, calculate the influence factor matrix of each parameter on the global geometric topology of the model, and then encapsulate each parameter of the ship structure model into parameter-constraint pairs based on the influence factor matrix.

[0015] Specifically, the initial parameters input during the ship structure design phase are obtained, and all initial parameters undergo deduplication and engineering value range verification. Duplicate parameters and abnormal parameters exceeding conventional engineering boundaries are eliminated. The remaining initial parameters are then integrated to form a candidate parameter set. Subsequently, a sensitivity scan is performed on the candidate parameter set. During the scan, each candidate parameter is sequentially selected as a perturbation variable, and a quantitative numerical perturbation is applied within its preset engineering allowable range. The global geometric topological response of the model caused by this perturbation is tracked throughout the process, and the corresponding topological change data is recorded. Based on the topological change data of each parameter, the comprehensive global topological impact value corresponding to each parameter is calculated, and the results are then processed according to parameter index and topology. The parameters are arranged according to their feature dimensions to construct an influence factor matrix for each parameter on the global geometric topology of the model. Based on the influence factor matrix, the comprehensive value of the global topological influence of each parameter is mapped to the probability level of model reconstruction failure. According to the preset engineering probability threshold, all candidate parameters are divided into hard parameters with strong topological correlation and flexible parameters with weak topological correlation. After classifying the parameter topological attributes, the corresponding geometric topological dependency path and correlation constraint conditions are matched for each parameter. At the same time, the asymmetric allowable fluctuation domain corresponding to the parameter is dynamically calculated in combination with the influence factor matrix. Then, the parameter value, geometric topological dependency path, and asymmetric allowable fluctuation domain are encapsulated into a unified parameter-constraint pair.

[0016] It should be further explained that the parameter-constraint pair is an independent encapsulated unit corresponding to a single parameter. The unit contains three types of information: parameter value, geometric topological dependency path, and asymmetric allowable fluctuation domain. Among them, the geometric topological dependency path records all topological features and subordinate connections associated with the parameter. The upper and lower limits of the asymmetric allowable fluctuation domain are dynamically mapped from the topological influence values ​​in the influence factor matrix. The fluctuation range of hard parameters with strong topological associations is narrower, while the fluctuation range of flexible parameters with weak topological associations is wider, thus balancing the stability of the topological structure and the flexibility of parameter adjustment.

[0017] It should be further explained that the specific formula for calculating the comprehensive value of global topology impact is as follows: ; In the formula, Indicates the first The combined global topological impact of each candidate parameter; Indicates the first Weight coefficients for each topological feature; This represents the total number of global geometric topological features in the model; Indicates the first After applying a unit perturbation to the parameters, the first... The amount of change that occurs in a topological feature.

[0018] The critical positioning module locates the critical equipotential surface that meets all requirements based on the numerical potential field and all parameter-constraint pairs, and at the same time calculates the minimum canonical distance vector of the current parameter-constraint pair from the critical equipotential surface.

[0019] Specifically, based on the numerical potential field constructed by the analytical construction module and all parameter-constraint pairs, a multi-dimensional parameter space coordinate system is constructed with each parameter as an independent dimension, corresponding one-to-one with the parameter-constraint pairs. The initial value of each parameter is mapped to the reference origin in the multi-dimensional parameter space coordinate system. Simultaneously, the allowable fluctuation domain carried in each parameter-constraint pair is mapped to the value range in the corresponding dimension. After completing the parameter space construction and boundary mapping, the overall sampling range is constrained within the value range of each parameter. A differentiated sampling strategy is executed based on the topological attributes of each parameter. Then, sampling is traversed by layering sampling points dimension by dimension to form a set of sampling points. The parameter combination corresponding to each sampling point is obtained and substituted into the numerical potential field to calculate the potential field value of that sampling point. When the potential field value of any sampling point is greater than or equal to the preset standard critical threshold, the sampling point is determined to be compliant; otherwise, it is determined to be non-compliant. After all sampling points have been calculated, they are divided into compliant and non-compliant point sets, forming discrete boundary point sets for compliant and non-compliant regions. Then, based on the discrete boundary point sets of the two types of regions, a spatial interpolation fitting algorithm is used to construct a continuous critical equipotential surface in the parameter space. During the fitting process, the module uses the gradient direction of the numerical potential field as the fitting guide to correct the surface of the hard parameters. After the critical equipotential surface is located, the shortest distance projection calculation is performed on the critical equipotential surface, starting from the reference origin corresponding to the current initial parameter combination, to obtain the minimum canonical distance vector from the reference origin to the critical equipotential surface. Then, based on the critical equipotential surface and the allowable fluctuation domain boundary of each parameter, the feasible parameter design domain range that simultaneously satisfies canonical constraints and topological feasibility is delineated. Finally, the critical equipotential surface, the minimum canonical distance vector, and the feasible parameter design domain range are sent to the parameter locking module.

[0020] It should be further explained that the differentiated sampling strategy is as follows: a high-density sampling step size is used for hard parameter dimensions, and a low-density sampling step size is used for flexible parameter dimensions, thereby reducing the amount of sampling calculation while ensuring the accuracy of the critical equipotential surface fitting.

[0021] It should be further explained that the specific formula for calculating the minimum normalized distance vector is as follows: ; In the formula, Represents the minimum normalized distance vector; Represents the coordinate vector of any point on the critical equipotential surface; This indicates the reference origin corresponding to the current combination of initial parameters; This represents the Euclidean norm.

[0022] Example 2, refer to Figure 1A ship structure design system based on parametric modeling includes an analytical construction module, a perception encapsulation module, a critical positioning module, a parameter locking module, a constraint propagation module, a model generation module, a topology verification module, and a specification verification module. The parameter locking module is used to calculate the optimal solution set of the design variable that meets the specifications based on the minimum specification distance vector when the user selects the target structural design variable. At the same time, the user can select values ​​within the range of the optimal solution set to obtain a set of compliant parameters.

[0023] Specifically, the system receives the critical equipotential surface, the minimum normalized distance vector, and the feasible region of parameter design. Simultaneously, it retrieves all parameter-constraint pairs and constructs a constraint framework for parameter locking operations, using the feasible region of parameter design as the value boundary, the critical equipotential surface as the compliance criterion, and the minimum normalized distance vector as the adjustment direction. Then, it obtains the user-selected target structural design variables. For each selected design variable, it matches its dimension and parameter-constraint pair in the parameter space, extracting the corresponding topological attributes, allowable fluctuation domain, and constraint information. It determines that the target structural design variables are within the feasible region of parameter design and obtains the user-selected optimization objective. When the user-selected optimization objective satisfies the preset specification, it follows the direction from the minimum normalized distance vector towards the critical equipotential surface, and then combines this with the feasible region boundary to solve for the critical value of the design variables. When the user selects the optimal margin as the optimization objective, the optimal margin range of the design variable is obtained by solving along the direction away from the critical equipotential surface from the minimum canonical distance vector and combining it with the feasible region boundary. The obtained value range is the optimal solution set of the corresponding target structure. Then, after the user selects the parameter value of the design variable within the optimal solution set, the parameter value is mapped to the corresponding coordinate point in the parameter space. Based on the geometric topological dependency path and constraint conditions carried in the parameter-constraint pair, the structural parameters associated with the design variable of the target structure are driven in reverse to perform linkage compensation. During the compensation process, the overall parameter combination is kept within the compliant region as a premise. The corresponding compensation amount is calculated in combination with the allowable fluctuation range of each associated parameter. After completing the linkage compensation of all associated parameters, a complete compliant parameter set is obtained.

[0024] It should be further explained that the optimization target is selected by the user through the interactive interface. The target satisfies the critical boundary of the standard corresponding to the preset standard, which can achieve the optimal parameter value under the premise of compliance. The preset optimal margin target corresponds to the safety margin range within the compliance area, which can reserve sufficient design redundancy on the basis of meeting the standard. The optimal solution set obtained under both types of targets falls completely within the feasible range of parameter design.

[0025] The constraint propagation module uses the compliance parameter set as the input basis. When the user modifies any structural parameter, it determines the adaptation strategy of downstream related features based on the topology path information in the parameter-constraint pair and completes the topology adaptive correction to obtain the complete parameter set.

[0026] Specifically, the compliance parameter set is used as the input basis for topology correction operations. Each parameter-constraint pair is used as an independent node in the graph, and the geometric topological dependency path carried in the parameter-constraint pair is used as the connecting edge of the graph to construct a bidirectional constraint graph. Simultaneously, user-initiated structural parameter modification commands are received, and the target node of the modified parameter in the bidirectional constraint graph is located. The topological attributes, allowable fluctuation domain, and path information of the corresponding parameter-constraint pair of the target node are obtained. It is determined whether the modified parameter value is within the allowable fluctuation domain of the parameter. If the modified parameter value is within the allowable fluctuation domain, the parameter value of the target node is updated. Starting from the updated target node, a full bidirectional traversal is performed along the connecting edge of the bidirectional constraint graph. During the traversal, all downstream feature nodes that are directly or indirectly related to the target node are sequentially retrieved along the topological dependency path. The parameters of each related node are simultaneously collected. The corresponding topological correlation strength, allowable fluctuation range of parameters, and current value information are used to form a complete downstream correlation feature set. Then, based on the downstream correlation feature set and combined with the topological attributes of each feature node, the adaptation strategy is determined one by one. If the feature node has a strong topological correlation, it is determined to be a forced following adaptation strategy; otherwise, it is determined to be a flexible adaptive strategy. For feature nodes using the forced following adaptation strategy, the corresponding linkage correction amount is calculated according to the modification amount of the target node and the topological correlation ratio. For feature nodes using the flexible adaptive strategy, the minimum topological distortion is used as the optimization objective, and the minimum energy offset is solved within the allowable fluctuation range. At the same time, the corrected value of each node is checked synchronously during the calculation. Then, the modified value of the target node, the corrected value of all associated nodes, and the remaining parameter values ​​that are not affected by this modification are integrated to form a complete parameter set.

[0027] The model generation module generates a three-dimensional geometric model of the ship structure based on a complete set of parameters and the topological path information in the parameter-constraint pairs, in accordance with the corresponding hierarchical order.

[0028] Specifically, based on the topology path information carried in the parameter-constraint pairs, a mapping relationship between each parameter and its corresponding geometric feature is established one by one. Simultaneously, according to the topology hierarchy attributes in the parameter-constraint pairs, a subset of parameters corresponding to the basic structure is extracted from the complete parameter set. Using a preset global reference coordinate system for the hull as the positioning reference, the geometric entity of the basic structure is generated. After the basic structure is generated, its reference plane and axis positions are fixed. Subsequently, based on the generated positioning reference of the basic structure, according to the component attachment relationships recorded in the topology path information, a subset of parameters corresponding to secondary components is extracted from the complete parameter set, and the geometric entities of the secondary structure components are generated sequentially. The primary structure is then assembled with secondary structural components at their corresponding positions in the primary structure. Based on the detailed subordination relationships in the topology path information, a subset of parameters corresponding to the detailed features is extracted from the complete parameter set. Detailed geometric features are then generated sequentially at the designated positions of the corresponding secondary components or primary structures. After all hierarchical geometric features are generated, they are integrated to form a complete three-dimensional geometric model of the ship structure. Then, based on the topology connection requirements in the parameter-constraint pairs, the connection status, fitting accuracy, and relative position between each component are checked one by one to identify any topological breaks or geometric interferences. After completing the topology consistency check, the final three-dimensional geometric model of the ship structure is obtained.

[0029] It should be further explained that the hierarchical order follows the conventional assembly logic of ship structural design. The basic structure serves as the global positioning benchmark, secondary components are attached to the layout of the basic structure, and detailed features are generated by attaching to the secondary components or the basic structure. The positioning relationship between each level is predefined by the topological path information in the parameter-constraint pairs.

[0030] The topology verification module is used to check the correspondence between the topological connections of geometric features and parameter constraints in the three-dimensional geometric model of the ship structure one by one, and to check whether there are geometric anomalies caused by parameter propagation.

[0031] The specification verification module is used to extract the parameters of each component from the three-dimensional geometric model of the ship structure, verify whether they meet all specification requirements one by one, and then generate a verification report and deviation item prompts based on the verification results.

Claims

1. A ship structure design system based on parametric modeling, characterized in that, It includes a parsing and construction module, a perception and encapsulation module, a critical localization module, a parameter locking module, a constraint propagation module, a model generation module, a topology verification module, and a specification verification module; The parsing construction module is used to decompose various structural design specification texts into parameter dependency logic trees, and to construct a numerical potential field based on each specification constraint in the parameter dependency logic tree. The perception encapsulation module is used to obtain the initial parameters of the ship structure, calculate the influence factor matrix of each parameter on the global geometric topology of the model, and then encapsulate each parameter of the ship structure model into parameter-constraint pairs based on the influence factor matrix. The critical positioning module locates the critical equipotential surface that meets all requirements based on the numerical potential field and all parameter-constraint pairs, and at the same time calculates the minimum canonical distance vector between the current parameter-constraint pair and the critical equipotential surface. The parameter locking module is used to calculate the optimal solution set of the design variable that meets the specifications based on the minimum specification distance vector when the user selects the target structural design variable. At the same time, the user can select values ​​within the range of the optimal solution set to obtain a set of compliant parameters. The constraint propagation module is used to take the compliance parameter set as the input basis. When the user modifies any structural parameter, it determines the adaptation strategy of the downstream associated features based on the topology path information in the parameter-constraint pair and completes the topology adaptive correction to obtain the complete parameter set. The model generation module generates a three-dimensional geometric model of the ship structure based on a complete set of parameters and according to the topological path information in the parameter-constraint pairs, in accordance with the corresponding hierarchical order. The topology verification module is used to check the correspondence between the topological connection relationship and parameter constraints of the geometric features in the three-dimensional geometric model of the ship structure one by one, and to check whether there are geometric anomalies caused by parameter propagation. The specification verification module is used to extract the parameters of each component from the three-dimensional geometric model of the ship structure, verify whether they meet all specification requirements one by one, and then generate a verification report and deviation item prompts based on the verification results.

2. The ship structure design system based on parametric modeling according to claim 1, characterized in that, The specific steps by which the perception encapsulation module encapsulates the structural parameters of the ship structure model into parameter-constraint pairs are as follows: S1.1: Obtain the initial parameters input during the ship structure design phase, and perform deduplication and engineering value range verification on all initial parameters. Eliminate duplicate definitions and abnormal parameters that exceed the conventional engineering boundaries. Then integrate the remaining initial parameters to form a candidate parameter set. Subsequently, perform sensitivity scanning on the candidate parameter set. During the scanning process, select each candidate parameter as a disturbance variable in turn, apply a quantitative numerical disturbance within its preset range allowed by engineering, and then track the global geometric topology response of the model caused by the disturbance throughout the process, while recording the corresponding topology change data. S1.2: Based on the topology change data of each parameter, calculate the comprehensive global topology impact value corresponding to each parameter, and arrange them according to the parameter index and topology feature dimension to construct the influence factor matrix of each parameter on the global geometric topology of the model. At the same time, based on the influence factor matrix, map the comprehensive global topology impact value of each parameter to the model reconstruction failure probability level, and divide all candidate parameters into hard parameters with strong topology correlation and flexible parameters with weak topology correlation according to the preset engineering probability threshold. S1.3: After completing the classification of parameter topology attributes, match the corresponding geometric topology dependency path and associated constraints for each parameter. At the same time, dynamically calculate the asymmetric allowable fluctuation domain corresponding to the parameter by combining the influence factor matrix. Then, encapsulate the parameter value, geometric topology dependency path and asymmetric allowable fluctuation domain into a unified parameter-constraint pair.

3. The ship structure design system based on parametric modeling according to claim 2, characterized in that, The specific steps for the critical positioning module to calculate the minimum canonical distance vector from the current parameter-constraint pair to the critical equipotential surface are as follows: S2.1: Based on the numerical potential field and all parameter-constraint pairs constructed by the analytical construction module, a multidimensional parameter space coordinate system is constructed with each parameter as an independent dimension, corresponding one-to-one with the parameter-constraint pairs. The initial value of each parameter is mapped to the reference origin in the multidimensional parameter space coordinate system. At the same time, the allowable fluctuation domain carried in each parameter-constraint pair is mapped to the value range in the corresponding dimension. After completing the parameter space construction and boundary mapping, the overall sampling range is constrained within the value range of each parameter. A differentiated sampling strategy is executed according to the topological attributes of each parameter. Then, sampling is traversed by layering points in each dimension to form a set of sampling points. S2.2: Obtain the parameter combination corresponding to each sampling point and substitute it into the numerical potential field to calculate the potential field value of the sampling point. When the potential field value of any sampling point is greater than or equal to the preset standard critical threshold, the sampling point is determined to be a compliant point; otherwise, it is determined to be a non-compliant point. After completing the calculation of all sampling points, all sampling points are divided into a compliant point set and a non-compliant point set to form a discrete boundary point set of compliant and non-compliant regions. Then, based on the discrete boundary point sets of the two types of regions, a spatial interpolation fitting algorithm is used to construct a continuous critical equipotential surface in the parameter space. During the fitting process, the module uses the gradient direction of the numerical potential field as the fitting guide to correct the surface of the hard parameters. S2.3: After the critical equipotential surface is located, the shortest distance projection calculation is performed on the critical equipotential surface, starting from the reference origin corresponding to the current initial parameter combination, to obtain the minimum gauge distance vector from the reference origin to the critical equipotential surface. Then, based on the boundary of the allowable fluctuation domain of the critical equipotential surface and each parameter, the feasible domain range of parameter design that simultaneously satisfies gauge constraints and topological feasibility is delineated. Finally, the critical equipotential surface, the minimum gauge distance vector, and the feasible domain range of parameter design are sent to the parameter locking module.

4. The ship structure design system based on parametric modeling according to claim 3, characterized in that, The specific steps by which the parameter locking module obtains the compliant parameter group are as follows: S3.1: Receive the critical equipotential surface, the minimum standard distance vector, and the feasible region of parameter design. Simultaneously, retrieve all parameter-constraint pairs and construct a constraint framework for parameter locking operations using the feasible region of parameter design as the value boundary, the critical equipotential surface as the compliance judgment benchmark, and the minimum standard distance vector as the adjustment direction basis. Then, obtain the target structural design variables selected by the user. For each selected design variable, match its dimension and parameter-constraint pairs in the parameter space, and obtain the corresponding topological attributes, allowed fluctuation domain, and constraint information from them to determine that the target structural design variables are within the feasible region of parameter design. S3.2: Obtain the user-selected optimization objective. When the user-selected optimization objective is to meet the preset specification, the critical value range of the design variable is obtained by following the direction of the minimum specification distance vector pointing to the critical equipotential surface and combining it with the feasible region boundary. When the user-selected optimization objective is the preset optimal margin, the optimal margin value range of the design variable is obtained by following the direction of the minimum specification distance vector away from the critical equipotential surface and combining it with the feasible region boundary. The value range obtained is the optimal solution set of the corresponding target structure. Then, after the user selects the parameter value of the design variable within the optimal solution set, the parameter value is mapped to the corresponding coordinate point in the parameter space. S3.3: Based on the geometric topology dependent path and constraint conditions carried in the parameter-constraint pair, the structural parameters that are related to the design variables of the target structure are reverse-driven to perform linkage compensation. During the compensation process, the overall parameter combination is kept within the compliance range. The corresponding compensation amount is calculated in combination with the allowable fluctuation range of each related parameter. After completing the linkage compensation of all related parameters, a complete compliant parameter set is obtained.

5. A ship structure design system based on parametric modeling according to claim 4, characterized in that, The specific steps by which the constraint propagation module obtains the complete parameter set are as follows: S4.1: Using the compliant parameter set as the input basis for topology correction operation, each parameter-constraint pair as an independent node of the graph, and the geometric topology dependency path carried in the parameter-constraint pair as the connection edge of the graph, a bidirectional constraint graph is constructed. At the same time, the user-initiated structural parameter modification command is received, the target node of the modified parameter in the bidirectional constraint graph is located, and the topology attribute, allowable fluctuation domain and path information of the parameter-constraint pair corresponding to the target node are obtained. It is determined whether the modified parameter value is within the allowable fluctuation domain of the parameter. If the modified parameter value is within the allowable fluctuation domain, the parameter value of the target node is updated. S4.2: Starting from the updated target node, perform a full-path bidirectional traversal along the connecting edges of the bidirectional constraint graph. During the traversal, sequentially retrieve all downstream feature nodes that are directly or indirectly related to the target node along the topological dependency path. Simultaneously collect the topological association strength, parameter allowable fluctuation range, and current value information of each associated node to form a complete downstream association feature set. Then, based on the downstream association feature set and combined with the topological attributes of each feature node, determine the adaptation strategy one by one. If it is a feature node with strong topological association, it is determined to be a forced following adaptation strategy; otherwise, it is determined to be a flexible adaptive strategy. S4.3: For feature nodes using the forced follow-adaptation strategy, the corresponding linkage correction amount is calculated according to the modification amount of the target node and the topology association ratio; for feature nodes using the flexible adaptive strategy, the minimum topology distortion is used as the optimization objective, and the minimum energy offset is solved within the allowable fluctuation domain. At the same time, the corrected value of each node is verified synchronously during the calculation process. Then, the modified value of the target node, the corrected value of all associated nodes, and the remaining parameter values ​​that are not affected by this modification are integrated to form a complete parameter set.

6. A ship structure design system based on parametric modeling according to claim 5, characterized in that, The specific steps of the model generation module in generating the three-dimensional geometric model of the ship structure according to the corresponding hierarchical order are as follows: S5.1: Based on the topology path information carried in the parameter-constraint pair, establish the mapping relationship between each parameter and the corresponding geometric feature one by one. At the same time, according to the topology hierarchy attribute in the parameter-constraint pair, extract the parameter subset corresponding to the basic structure from the complete parameter set, and generate the geometric entity of the basic structure with the preset global reference coordinate system of the hull as the positioning reference. After the basic structure is generated, fix its reference plane and axis position. Then, on the generated basic structure positioning reference, according to the component attachment relationship recorded in the topology path information, extract the parameter subset corresponding to the secondary component from the complete parameter set, generate the geometric entity of the secondary structure component in sequence, and then assemble the secondary structure component to the corresponding position of the basic structure. S5.2: Based on the detailed subordination relationships in the topology path information, extract the parameter subsets corresponding to the detailed features from the complete parameter set, and generate detailed geometric features sequentially at the specified positions of the corresponding secondary components or basic structures. After all hierarchical geometric features are generated, they are integrated to form a complete three-dimensional geometric model of the ship structure. Then, based on the topology connection requirements in the parameter-constraint pairs, verify the connection status, fitting accuracy and relative position between each component one by one, check for topology breaks or geometric interference, and after completing the topology consistency verification, obtain the final three-dimensional geometric model of the ship structure.

Images