Multi-target collaborative simulation optimization system and method for ship structure

Abstract

The invention provides a ship structure-oriented multi-target collaborative simulation optimization system and method, and relates to the technical field of simulation optimization, and the method comprises the steps: constructing a ship structure model; generating a multi-simulation optimization target; configuring an operation model and a finite element analysis model; constructing a simulation space; finite element analysis model parameter adjustment is executed, simulation testing is carried out, and simulation testing data are output; calculating the simulation test data to obtain a test calculation result, and performing anomaly identification on the test calculation result to obtain an abnormal parameter; and carrying out analysis positioning, determining an optimization variable, and carrying out optimization on the optimization variable through a preset optimization algorithm to obtain ship structure optimization information. According to the method and the device, the technical problem of low parameter optimization accuracy and simulation efficiency due to the fact that most ship structure simulation does not have accurate simulation test and data analysis in the prior art is solved, the technical goal of improving the accuracy of simulation test and data analysis is achieved, and the technical effect of improving the simulation efficiency and accuracy is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of simulation optimization, and particularly relates to a multi-objective cooperative simulation optimization system and method for ship structure. BACKGROUND

[0002] The simulation of ship structure refers to using computer simulation technology to analyze and predict the behavior and performance of ship structure under actual working conditions.

[0003] At present, in the existing simulation process of ship structure, the ship structure design is mostly not accurately simulated, tested and analyzed, and abnormal parameters are not identified, so that the demand of actual working conditions cannot be met, or unnecessary cost and weight are increased, and the abnormal parameters cannot be identified, which may cause safety risks in actual operation of the ship.

[0004] In summary, in the prior art, since the simulation of ship structure mostly does not have accurate simulation test and data analysis, the parameter optimization accuracy is low, which causes low simulation efficiency and further affects the simulation effect of ship structure. SUMMARY

[0005] The purpose of the present application is to provide a multi-objective cooperative simulation optimization system and method for ship structure, which solves the technical problem in the prior art that since the simulation of ship structure mostly does not have accurate simulation test and data analysis, the parameter optimization accuracy is low, which causes low simulation efficiency and further affects the simulation effect of ship structure.

[0006] In view of the above problems, the present application provides a multi-objective cooperative simulation optimization system and method for ship structure.

[0007] In a first aspect, the application also provides a ship structure-oriented multi-objective cooperative simulation optimization system for executing a ship structure-oriented multi-objective cooperative simulation optimization method, wherein the system comprises: a ship structure model construction module for performing structure geometric parameterization based on the partition characteristics of the ship physical structure to construct a ship structure model; a simulation optimization target acquisition module for analyzing the coupling relationship of the expected function, performance index and application scenario to generate multi-simulation optimization targets, the multi-simulation optimization targets corresponding to the targets of the expected function, performance index and application scenario, respectively; a model configuration module for configuring an operation model and a finite element analysis model according to the simulation optimization targets; a simulation space construction module for coupling the ship structure model, the finite element analysis model and the operation model into a unified simulation space through a parameter mapping interface; a simulation test data output module for performing adaptive step adjustment of the parameters of the finite element analysis model within the parameter range of the expected function, performance index and application scenario, and outputting simulation test data based on the adjusted parameters; an abnormal parameter acquisition module for performing operation on the simulation test data through the operation model to obtain test operation results, and performing abnormal identification on the test operation results based on the simulation optimization targets to obtain abnormal parameters; and an optimization variable determination module for determining optimization variables based on the abnormal parameters, and performing optimization on the optimization variables through a preset optimization algorithm to obtain ship structure optimization information.

[0008] In a second aspect, the application provides a ship structure-oriented multi-objective cooperative simulation optimization method, which is implemented by a ship structure-oriented multi-objective cooperative simulation optimization system, wherein the method comprises: performing structure geometric parameterization based on the partition characteristics of the ship physical structure to construct a ship structure model; analyzing the coupling relationship of the expected function, performance index and application scenario to generate multi-simulation optimization targets, the multi-simulation optimization targets corresponding to the targets of the expected function, performance index and application scenario, respectively; configuring an operation model and a finite element analysis model according to the simulation optimization targets; coupling the ship structure model, the finite element analysis model and the operation model into a unified simulation space through a parameter mapping interface; performing adaptive step adjustment of the parameters of the finite element analysis model within the parameter range of the expected function, performance index and application scenario, and outputting simulation test data based on the adjusted parameters; performing operation on the simulation test data through the operation model to obtain test operation results, and performing abnormal identification on the test operation results based on the simulation optimization targets to obtain abnormal parameters; determining optimization variables based on the abnormal parameters, and performing optimization on the optimization variables through a preset optimization algorithm to obtain ship structure optimization information.

[0009] The one or more technical solutions provided in the application have at least the following technical effects or advantages: By utilizing the partitioned features of the ship's physical structure, structural geometric parameterization is performed to construct a ship structural model. The coupling relationship between expected functions, performance indicators, and application scenarios is analyzed to generate multiple simulation optimization objectives, each corresponding to one of the expected functions, performance indicators, and application scenarios. A computational model and a finite element analysis model are configured according to these optimization objectives. The ship structural model, finite element analysis model, and computational model are coupled into a unified simulation space through a parameter mapping interface. Within the parameter range of the expected functions, performance indicators, and application scenarios, adaptive step-size adjustments are performed on the parameters of the finite element analysis model. Simulation tests are conducted based on these adjusted parameters to output simulation test data. The computational model is used to perform calculations on the simulation test data to obtain test results. Anomalies in the test results are identified based on the simulation optimization objectives to obtain abnormal parameters. These abnormal parameters are analyzed and located to determine optimization variables. A preset optimization algorithm is used to optimize these variables, obtaining ship structural optimization information. This achieves the technical goal of improving the accuracy of simulation testing and data analysis, and enhances simulation efficiency and accuracy.

[0010] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of the multi-objective collaborative simulation optimization system for ship structures in this application; Figure 2 This is a flowchart illustrating the multi-objective collaborative simulation optimization method for ship structures proposed in this application.

[0013] Explanation of reference numerals in the attached figures: Ship structure model construction module 11, simulation optimization target acquisition module 12, model configuration module 13, simulation space construction module 14, simulation test data output module 15, abnormal parameter acquisition module 16, and optimization variable determination module 17. Detailed Implementation

[0014] This application provides a multi-objective collaborative simulation optimization system and method for ship structures, solving the technical problem in existing technologies where the lack of accurate simulation testing and data analysis in ship structure simulations leads to low accuracy in parameter optimization, resulting in low simulation efficiency and further affecting the simulation effect. It achieves the technical goal of improving the accuracy of simulation testing and data analysis, thus enhancing both simulation efficiency and accuracy.

[0015] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0016] Example 1 This application provides a multi-objective collaborative simulation and optimization system for ship structures. Please refer to the appendix. Figure 1 The system includes: Ship structure model building module 11 is used to perform structural geometric parameterization based on the partition characteristics of the ship's physical structure and build a ship structure model; The simulation optimization target acquisition module 12 is used to analyze the coupling relationship between expected functions, performance indicators and application scenarios, and generate multiple simulation optimization targets, which correspond to the targets of expected functions, performance indicators and application scenarios respectively. Model configuration module 13 is used to configure the computational model and finite element analysis model according to the simulation optimization target; Simulation space construction module 14 is used to couple the ship structure model, finite element analysis model and calculation model into a unified simulation space through a parameter mapping interface; The simulation test data output module 15 is used to perform adaptive step size adjustment of the parameters of the finite element analysis model within the parameter range of the expected function, performance index and application scenario, and output simulation test data based on the adjusted parameters. The abnormal parameter acquisition module 16 is used to perform calculations on the simulation test data through the calculation model to obtain test calculation results, and to perform anomaly identification on the test calculation results based on the simulation optimization target to obtain abnormal parameters. The optimization variable determination module 17 is used to analyze and locate the abnormal parameters, determine the optimization variables, and optimize the optimization variables through a preset optimization algorithm to obtain ship structure optimization information.

[0017] Furthermore, the model configuration module 13 in the system is also used for: Based on the expected functions, performance indicators, and application scenarios, the parameters are parsed, and the correlation analysis is performed on the parsed parameters to construct a parameter topology structure, which includes parameter relationships and relationship coefficients. Based on the expected functions, performance indicators, and application scenarios, target evaluation parameters are set. Using the target evaluation parameters as the search direction, relationship extraction is performed based on the parameter topology to construct a target evaluation parameter relationship set; Training data is constructed based on the target evaluation parameter relationship set, and the model is trained using the training data until the target is converged to obtain the computational model, which has target evaluation parameter identification information.

[0018] Furthermore, the model configuration module 13 in the system is also used for: Based on the ship's physical structure, a coarse-grained grid is set. The coarse-grained grid is used to determine the grid division rules based on the ship's hull functional partitions, unit component partitions, and shape structure partitions of the ship's physical structure. Based on the simulation optimization objective, the target hull structure is determined, wherein the target hull structure is the hull structure that has an impact on the simulation optimization objective; An impact analysis is performed on the target hull structure, and a fine-grained mesh is set based on the impact. The size of the fine-grained mesh is inversely proportional to the impact; the greater the impact, the smaller the mesh. Define the properties of ship materials, including elastic modulus, thermal conductivity, density, and hardness; A finite element analysis model is constructed based on the coarse-grained mesh, fine-grained mesh, and ship material properties. The loads and boundary conditions are determined according to the simulation optimization objectives, and the loads and boundary conditions are fitted to the finite element analysis model.

[0019] Furthermore, the simulation test data output module 15 in the system is also used for: Based on the parameter range, set the test adjustment step size; Based on the test adjustment step size and parameter range, the parameters are adjusted, and the stress change is decomposed according to the coarse-grained grid and fine-grained grid to obtain the stress value change density of each grid. When the stress value change density reaches a preset threshold, the mesh parameters are adjusted based on the stress value gradient to determine the mesh configuration parameters. Based on the parameter range and test adjustment step size, the load parameters are adjusted to determine the load configuration parameters. The parameters of the finite element analysis model are then configured using the mesh configuration parameters and load configuration parameters.

[0020] Furthermore, the abnormal parameter acquisition module 16 in the system is also used for: Based on the aforementioned simulation optimization objectives, an abnormal case record dataset is obtained; Anomaly decomposition is performed based on the dataset of abnormal case records to determine the abnormal factors and corresponding abnormal thresholds. The abnormal factors are those that affect the abnormal case events. Based on the aforementioned abnormal factors and abnormal thresholds, an abnormal identification module is configured to identify abnormal factors in the test operation results and to determine abnormalities through the abnormal thresholds to obtain abnormal parameters, wherein the abnormal parameters are abnormal factors that exceed the abnormal thresholds.

[0021] Furthermore, the optimization variable determination module 17 in the system is also used for: Based on the aforementioned abnormal case record dataset, abnormal time series data decomposition is performed to construct time series training data; The time series prediction model is trained using the aforementioned time series training data to obtain an anomaly prediction model; The abnormal parameters are introduced into the anomaly prediction model. Starting from the result of the abnormal event risk, reverse parameter tracking is performed based on the anomaly prediction model, and the impact of parameter changes on the output result is quantified. The optimization variables are selected based on their impact, and the optimization variables are ship structural parameters whose impact is greater than a threshold.

[0022] Furthermore, the abnormal parameter acquisition module 16 in the system is also used for: When the abnormal parameter output is empty, an abnormal prediction is performed based on the test calculation results using the abnormal prediction model to obtain the abnormal prediction probability. Based on the anomaly prediction probability, risk events are screened, and the risk events are anomaly events whose anomaly prediction probability reaches a preset requirement; Based on the risk event and the test calculation results, the abnormal parameter is determined. The abnormal parameter is the abnormal factor corresponding to the risk event, and the difference between the abnormal factor and the abnormal threshold in the test calculation results is less than a preset condition.

[0023] Furthermore, the optimization variable determination module 17 in the system is also used for: Based on the optimization variables, a random population is established, and the random population is divided into a first population and a second population. Based on the first population and the second population, determine the optimization center and optimization direction; Based on the optimization variables, an analysis of the correlation and impact on the ship's physical structure is conducted to determine the correlation structural parameters and the linkage effect. Based on the associated structural parameters and the linkage influence, determine the constraints of the optimization variables; Based on the constraints, a search is performed using the optimization center and optimization direction, the search results are evaluated using a fitness function, and the population is updated based on the fitness. When the update frequency does not meet the update iteration requirements, the optimal solution of the current population is added to the taboo space, and the optimization is performed iteratively based on the neighborhood space of the current population. The update frequency judgment result is used as the taboo condition. When the optimization target or number of iterations is reached, the current optimal solution and the optimal solution in the tabu space are obtained to evaluate fitness and determine the ship structure optimization information.

[0024] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. The multi-objective collaborative simulation optimization method and specific examples for ship structures are also applicable to the multi-objective collaborative simulation optimization system for ship structures in this embodiment. Through the detailed description of the multi-objective collaborative simulation optimization method for ship structures, those skilled in the art can clearly understand the multi-objective collaborative simulation optimization system for ship structures in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As for the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and relevant parts can be referred to in the method section.

[0025] Example 2 Please see the appendix Figure 2 This application provides a multi-objective collaborative simulation optimization method for ship structures, wherein the method is applied to a multi-objective collaborative simulation optimization system for ship structures, and the method specifically includes the following steps: Step 1: Based on the partitioning characteristics of the ship's physical structure, perform structural geometric parameterization and construct a ship structural model.

[0026] Specifically, modern measurement technologies, such as laser scanning and 3D photogrammetry, are used to accurately acquire physical structural data of the ship. CAD software is then used to construct a 3D structural model of the ship based on the acquired data, ensuring that the model accurately reflects the ship's actual dimensions and design features.

[0027] Step 2: Analyze the coupling relationship between expected functions, performance indicators and application scenarios, and generate multiple simulation optimization targets. The multiple simulation optimization targets correspond to the expected functions, performance indicators and application scenarios respectively.

[0028] Specifically, the main functions that the ship design must meet should be clearly defined, such as load capacity, stability, and seakeeping. Performance indicators for evaluating the design's merits should be established, such as hull strength, fatigue life, and hydrodynamic performance. Based on the ship's expected application environment, such as different sea states and cargo conditions, the design should ensure good performance under various application scenarios. A simulation model should be built based on the 3D model, and the model should be meshed to ensure the mesh quality meets the simulation accuracy requirements. Based on the simulation optimization objectives, an appropriate optimization algorithm should be selected, such as genetic algorithm, particle swarm optimization, or simulated annealing. The simulation model and optimization algorithm should be combined for iterative optimization to find the optimal design scheme that satisfies all performance indicators.

[0029] Step 3: Configure the computational model and finite element analysis model according to the simulation optimization objectives.

[0030] Specifically, based on the simulation optimization objectives, appropriate computational models, including machine learning models, optimization algorithms, or other mathematical models, are selected or developed to process and analyze simulation data. The finite element analysis model is then configured according to the simulation optimization objectives, including determining the analysis type, mesh generation, and material properties. This ensures that the finite element model can accurately simulate the physical behavior of the ship structure.

[0031] Step 4: Couple the ship structure model, finite element analysis model, and computation model into a unified simulation space through the parameter mapping interface.

[0032] Specifically, the ship structural model is combined with the finite element analysis model to ensure that the finite element model accurately reflects the structural characteristics of the ship. The computational model is combined with the finite element model to enable efficient processing and analysis of the simulation results. The ship structural model, finite element analysis model, and computational model are fitted together to construct a comprehensive simulation space. This simulation space is a virtual environment capable of simulating the behavior of the ship under actual operating conditions and allows designers to conduct various simulation experiments.

[0033] Step 5: Within the parameter range of the expected function, performance indicators and application scenario, perform adaptive step size adjustment of the finite element analysis model parameters, and output simulation test data based on the adjusted parameters.

[0034] Specifically, based on the expected functions, performance indicators, and application scenarios, the configuration parameters of the finite element analysis model are adjusted, including material properties, boundary conditions, and load conditions. Simulation tests are then conducted using the adjusted configuration parameters. The simulation test output includes data such as stress, deformation, and frequency, reflecting the performance of the ship structure under specific conditions.

[0035] Step Six: Perform calculations on the simulation test data using the calculation model to obtain test calculation results. Based on the simulation optimization objective, identify anomalies in the test calculation results to obtain abnormal parameters.

[0036] Specifically, simulation test data is input into the computational model. The data is processed and analyzed to obtain the test results. Based on the simulation optimization objectives, anomalies in the test results are identified to pinpoint parameters that cause structural performance degradation or failure, and then optimized accordingly.

[0037] Step 7: Analyze and locate the abnormal parameters to determine the optimization variables, and optimize the optimization variables using a preset optimization algorithm to obtain ship structure optimization information.

[0038] Specifically, in-depth analysis of anomalous parameters, pinpointing their exact location and impact within the ship's structure, helps to understand their influence on the overall structural performance. Based on the analysis results, optimization variables are determined. These are design parameters that significantly affect structural performance and can be adjusted. Pre-defined optimization algorithms, including genetic algorithms, particle swarm optimization, and simulated annealing, are used to optimize these variables. Through this optimization process, optimized information about the ship's structure is obtained, and design parameters are adjusted to improve structural performance, thereby meeting the simulation optimization objectives.

[0039] The proposed multi-objective collaborative simulation optimization method for ship structures is applied to a multi-objective collaborative simulation optimization system for ship structures, achieving the technical goal of improving the accuracy of simulation testing and data analysis, and thus improving the technical effect of simulation efficiency and accuracy.

[0040] Furthermore, based on the simulation optimization target configuration computation model, this application also includes: Parameters are analyzed according to the expected functions, performance indicators, and application scenarios, and correlation analysis is performed on the analyzed parameters to construct a parameter topology structure, which includes parameter relationships and relationship coefficients. Target evaluation parameters are set based on the expected functions, performance indicators, and application scenarios. Using the target evaluation parameters as the search direction, relationships are extracted based on the parameter topology structure to construct a target evaluation parameter relationship set. Training data is constructed based on the target evaluation parameter relationship set, and the model is trained using the training data until the target converges, obtaining the computational model, which has target evaluation parameter identification information.

[0041] Specifically, a detailed analysis is conducted on each parameter in the ship design, including hull dimensions, material properties, and structural layout. Based on the expected function, performance indicators, and application scenarios, the impact and importance of each parameter on the design are determined. Statistical methods or data mining techniques are used to analyze the correlations between parameters to identify those that significantly affect ship performance, helping to understand the interactions between parameters and providing a basis for subsequent optimization work. Based on the correlation analysis results, a graph of relationships between parameters is constructed, i.e., the parameter topology. Each parameter is treated as a node, and the relationships and coefficients between parameters are treated as edges.

[0042] Then, based on the design goals and performance indicators, key parameters are selected from all parameters as target evaluation parameters, serving as the main focus during the optimization process.

[0043] Next, based on the parameter topology, the relationships related to the target evaluation parameters are extracted, a target evaluation parameter relationship set is constructed, and the target evaluation parameters are optimized by adjusting the relevant parameters.

[0044] Next, data is generated or collected using the simulation model to construct a dataset for training the machine learning model, containing input parameters and corresponding output performance metrics. The machine learning model, such as a neural network, support vector machine, or other suitable model, is then trained using the training data. The training process requires multiple iterations until the model reaches convergence, meaning its predictive performance no longer significantly improves. After training, the resulting computational model can predict ship performance based on the input parameters.

[0045] By optimizing the target configuration computational model through simulation, potential problems can be identified during the design phase, reducing the number of times physical prototypes need to be built and tested, thereby shortening the design cycle.

[0046] Furthermore, based on the finite element analysis model configured according to the simulation optimization objective, this application also includes: Based on the ship's physical structure, a coarse-grained mesh is defined. This coarse-grained mesh is used to determine the mesh segmentation rules based on the ship's hull functional partitioning, unit component partitioning, and shape structure partitioning. Based on the simulation optimization objective, a target hull structure is determined. This target hull structure is one that has an impact on the simulation optimization objective. An impact analysis is performed on the target hull structure, and a fine-grained mesh is defined based on this impact. The size of the fine-grained mesh is inversely proportional to the impact; the greater the impact, the smaller the mesh size. Ship material properties are defined, including elastic modulus, thermal conductivity, density, and hardness. A finite element analysis model is constructed based on the coarse-grained mesh, fine-grained mesh, and ship material properties. Loads and boundary conditions are determined based on the simulation optimization objective, and these loads and boundary conditions are fitted to the finite element analysis model.

[0047] Specifically, the simulation optimization objective involves configuring a finite element analysis model. Based on the simulation optimization objective, a suitable finite element analysis software is selected, and the basic parameters of the model are configured. The analysis type is then determined, such as static analysis, dynamic analysis, thermal analysis, or fluid dynamics analysis.

[0048] Then, based on the ship's physical structure, the hull is divided into different functional zones, unit component zones, and shape structure zones. According to the zoning rules, larger grid points are set for parts that are less related to the optimization objective, and smaller grid points are set for parts that are more related to the optimization objective, in order to adapt to the large-scale structure and function of the ship.

[0049] Next, identify the hull structure components that significantly impact the simulation optimization objectives, including stress concentration areas, critical support structures, or parts that decisively affect ship performance. Conduct a detailed impact analysis of the target hull structure to determine the degree of influence on overall performance, including stress analysis, deformation analysis, and frequency analysis.

[0050] Next, based on the results of the impact analysis of the target hull structure, a finer mesh is used for high-impact areas. Fine-grained meshes provide higher resolution, thus capturing key information such as stress concentration and complex deformations more accurately.

[0051] Then, the physical properties of the ship materials, such as elastic modulus, thermal conductivity, density, and hardness, are input to simulate the material's behavior under actual working conditions.

[0052] Next, combining coarse-grained and fine-grained meshes with defined material properties, a complete finite element model is constructed, ensuring the mesh quality meets analysis requirements and avoiding over-meshing or mesh distortion. Based on the simulation optimization objectives, loads acting on the hull, such as gravity, buoyancy, and wave loads, are determined. Boundary conditions, such as fixation constraints and displacement constraints, are set to simulate the actual working environment. The determined loads and boundary conditions are then applied to the finite element model. It is ensured that the application of loads and boundary conditions conforms to actual conditions to obtain accurate analysis results.

[0053] By constructing accurate finite element models, we can simulate the performance of ships under actual working conditions, evaluate and optimize the design of ship structures, and ensure that predetermined functions, performance indicators and application scenarios are met.

[0054] Furthermore, based on the parameter range of the expected function, performance indicators, and application scenario, this application also includes: performing adaptive step size adjustment of the finite element analysis model parameters; Based on the parameter range, a test adjustment step size is set; parameters are adjusted based on the test adjustment step size and parameter range; stress variation is decomposed according to the coarse-grained and fine-grained grids to obtain the stress value variation density of each grid; when the stress value variation density reaches a preset threshold, grid parameters are adjusted based on the stress value gradient to determine the grid configuration parameters; load parameters are adjusted according to the parameter range and test adjustment step size to determine the load configuration parameters; and the parameters of the finite element analysis model are configured using the grid configuration parameters and load configuration parameters.

[0055] Specifically, the step size for parameter adjustment is determined based on the parameter range. The test adjustment step size is the increment during parameter adjustment, which determines the fineness of parameter changes.

[0056] Then, within the parameter range, parameters are adjusted based on a set step size. After each parameter adjustment, the stress variation of the model is analyzed using coarse-grained and fine-grained meshes to obtain the stress variation density for each mesh. Stress changes significantly over short distances; high stress gradient regions, such as those at abrupt geometric changes, notches, holes, cracks, or interfaces between different materials, are often the starting points for structural failure. Refining the mesh is crucial to achieving accurate simulation results. The stress variation density reflects the impact of parameter changes on stress distribution.

[0057] Next, when the stress change density of the mesh reaches a preset threshold, it indicates that the parameters have a significant impact on the stress distribution; conversely, the impact is weaker. The mesh parameters are adjusted based on the stress gradient, i.e., the rate of stress change, to optimize the design. Mesh configuration parameters are then determined to guide further mesh refinement or coarsening.

[0058] Next, the load parameters are adjusted to observe and optimize their impact on model performance. When adjusting load parameters, the set parameter range and test adjustment step size must be followed. The finite element analysis model is configured using mesh configuration parameters and load configuration parameters. Ensure that the model's parameter settings accurately simulate the behavior of the ship under actual operating conditions.

[0059] By optimizing the parameter settings of the finite element analysis model, the accuracy and efficiency of the analysis can be improved, gradually approaching the optimal design, while reducing the need for physical testing, thereby saving time.

[0060] Furthermore, based on the simulation optimization objective, anomaly identification is performed on the test calculation results to obtain abnormal parameters. This application also includes: Based on the simulation optimization objective, an abnormal case record dataset is obtained; anomaly decomposition is performed on the abnormal case record dataset to determine abnormal factors and corresponding abnormal thresholds, wherein the abnormal factors are factors that affect abnormal case events; based on the abnormal factors and abnormal thresholds, an anomaly identification module is configured to identify abnormal factors in the test operation results, and anomaly discrimination is performed through the abnormal thresholds to obtain abnormal parameters, wherein the abnormal parameters are abnormal factors that exceed the abnormal thresholds.

[0061] Specifically, based on the simulation optimization objectives, record data of abnormal cases over a historical period are collected to obtain an abnormal case record dataset, including ship design, operating conditions, environmental factors, material properties, and related abnormal events.

[0062] Then, the dataset of abnormal case records is analyzed to identify the main factors leading to the abnormal events, and to determine the abnormal factors and thresholds. Abnormal factors may include design parameters, environmental conditions, operational errors, etc., and have a significant impact on the occurrence of abnormal cases.

[0063] Next, based on the determined abnormal factors and abnormal thresholds, an anomaly identification module is configured to identify abnormal factors in the test calculation results. The anomaly identification module analyzes the test calculation results, comparing the results with the abnormal thresholds to determine if any abnormal factors exceed the thresholds. Factors exceeding the abnormal thresholds are marked as abnormal parameters, and the parts that need adjustment are obtained.

[0064] By identifying potential anomalies in a timely manner during the simulation optimization process, similar problems can be avoided in actual operation, which helps to improve the reliability and safety of the design and reduce the risk of accidents.

[0065] Furthermore, based on the abnormal parameters, the application analyzes and locates the anomaly to determine the optimization variable, and also includes: Based on the aforementioned abnormal case record dataset, abnormal time-series data decomposition is performed to construct time-series training data. The time-series training data is used to train a time-series prediction model to obtain an abnormal prediction model. The abnormal parameters are introduced into the abnormal prediction model. Starting from the result of abnormal event risk, reverse parameter tracking is performed based on the abnormal prediction model, and the impact of parameter changes on the output result is quantified. The optimization variables are selected according to the impact, and the optimization variables are ship structural parameters with an impact greater than a threshold.

[0066] Specifically, time-series data analysis is performed on the abnormal case record dataset to identify the time-series characteristics of abnormal events, which are used to construct a training dataset containing time-series data and corresponding abnormal labels.

[0067] Then, the time series prediction model is trained using time series training data. Analytical methods such as Long Short-Term Memory networks or gated recurrent units are employed. After training, an anomaly prediction model is obtained, capable of predicting the risk of anomalous events based on the input time series data.

[0068] Next, the identified anomalous parameters are incorporated into the anomaly prediction model. Starting with the results of the anomalous event risk assessment, the anomaly prediction model is used for reverse parameter tracking to identify the parameters that lead to the occurrence of the anomalous event. The impact of parameter changes on the model output is analyzed, and by quantifying the impact, the contribution of each parameter to the occurrence of the anomalous event is obtained.

[0069] Next, based on the results of the quantitative analysis, ship structural parameters with an impact greater than a threshold were selected as optimization variables. Optimization variables are parameters that require focused attention and adjustment during the design optimization process.

[0070] By using models to predict the risk of anomalous events and identifying and quantifying key factors leading to anomalies through inverse parameter tracking, potential problems can be identified and addressed during the design phase, thereby improving the reliability and safety of the design. This ensures the accuracy and predictive power of the model.

[0071] Furthermore, the method for obtaining abnormal parameters also includes: When the output of the abnormal parameter is empty, anomaly prediction is performed through the anomaly prediction model based on the test calculation results to obtain the anomaly prediction probability; based on the anomaly prediction probability, risk events are screened, and the risk events are anomalies whose anomaly prediction probability reaches a preset requirement; based on the risk events and the test calculation results, the abnormal parameter is determined, and the abnormal parameter is the anomaly factor corresponding to the risk event, and the difference between the abnormal factor and the anomaly threshold in the test calculation results is less than a preset condition.

[0072] Specifically, when no anomalous parameters are identified, it means that no factors exceeding the anomalous threshold were directly found in the test results. Instead, an anomaly prediction model is used to assess potential anomaly risks. The anomaly prediction model is used to analyze the test results to predict the likelihood of future anomalous events. The model outputs a probability value, representing the probability of an anomaly occurring given the parameters and conditions.

[0073] Then, based on the predicted probability of anomalies, anomalies that meet a preset probability threshold are selected as risk events. The preset threshold is a probability threshold; anomalies exceeding this threshold are considered high-risk.

[0074] Next, for the selected risk events, the abnormal parameters leading to the risk events are determined based on the test results. Abnormal parameters are anomalous factors in the test results whose difference from the abnormal threshold is less than a preset condition. These parameters do not directly manifest as anomalies in the current test, but according to the model's predictions, they may lead to abnormal events in the future.

[0075] By utilizing anomaly prediction models to identify and assess potential anomaly risks, key factors that may lead to anomalies can be discovered even without direct anomaly parameter outputs. This helps to provide early warning and prevention of potential anomaly events, thereby improving the reliability and safety of the design.

[0076] Furthermore, the method of optimizing the optimization variables using a preset optimization algorithm to obtain ship structure optimization information also includes: Based on the optimization variables, a random population is established and divided into a first population and a second population. Based on the first and second populations, an optimization center and optimization direction are determined. An association influence analysis is performed on the ship's physical structure based on the optimization variables to determine associated structural parameters and linkage influence. Based on the associated structural parameters and linkage influence, constraints on the optimization variables are determined. Based on the constraints, a search is performed using the optimization center and optimization direction. The search results are evaluated using a fitness function, and the population is updated based on the fitness. When the update frequency does not meet the update iteration requirements, the optimal solution of the current population is added to the taboo space. Iterative optimization is performed based on the neighborhood space of the current population, and the update frequency judgment result is used as a taboo condition. When the optimization objective or iteration number is reached, the current optimal solution and the optimal solution in the taboo space are obtained for fitness evaluation to determine ship structure optimization information.

[0077] Specifically, based on the optimization variables, a random population is initialized, where each individual represents a potential design solution. The random population is then divided into two groups, a first group and a second group, and different strategies or algorithms are used to handle different subproblems.

[0078] Then, based on the information from the first and second populations, the optimization center is determined, which is the approximate location of the potential optimal solution in the population. The optimization direction is determined, which is the direction in which individuals in the population move towards a better solution.

[0079] Next, an impact analysis of the ship's physical structure is conducted to identify interrelated structural parameters and their synergistic effects, and to obtain the interactions between these parameters for subsequent optimization.

[0080] Next, based on the associated structural parameters and the linkage effect, the constraints of the optimization variables are determined to ensure that the optimization process does not produce unrealistic or unsafe design solutions.

[0081] Then, under constraints, a search is conducted using the optimization center and optimization direction to find a better design solution. A fitness function is used to evaluate the search results, measuring how well the design solution satisfies the optimization objective. Based on the fitness evaluation results, the population is updated, retaining better individuals and eliminating poorer ones.

[0082] Next, when the population update frequency no longer meets the iteration requirements, the optimal solution of the current population is added to the tabu space to avoid redundant searches. Iterative optimization is then performed within the neighborhood space of the current population to find a better design. The update frequency judgment result serves as a tabu condition to ensure the diversity of the search.

[0083] Next, when the optimization target or number of iterations is reached, the fitness of the current optimal solution is compared with that of the optimal solution in the taboo space to determine the optimal design scheme, i.e., the ship structure optimization information.

[0084] By effectively finding a ship structure design scheme that satisfies the optimization objective, it is possible to handle complex optimization problems and take into account the mutual influence and constraints between multiple optimization variables.

[0085] In summary, the multi-objective collaborative simulation optimization method for ship structures provided in this application has the following technical effects: By utilizing the partitioned features of the ship's physical structure, structural geometric parameterization is performed to construct a ship structural model. The coupling relationship between expected functions, performance indicators, and application scenarios is analyzed to generate multiple simulation optimization objectives, each corresponding to one of the expected functions, performance indicators, and application scenarios. A computational model and a finite element analysis model are configured according to these optimization objectives. The ship structural model, finite element analysis model, and computational model are coupled into a unified simulation space through a parameter mapping interface. Within the parameter range of the expected functions, performance indicators, and application scenarios, adaptive step-size adjustments are performed on the parameters of the finite element analysis model. Simulation tests are conducted based on these adjusted parameters to output simulation test data. The computational model is used to perform calculations on the simulation test data to obtain test results. Anomalies in the test results are identified based on the simulation optimization objectives to obtain abnormal parameters. These abnormal parameters are analyzed and located to determine optimization variables. A preset optimization algorithm is used to optimize these variables, obtaining ship structural optimization information. This achieves the technical goal of improving the accuracy of simulation testing and data analysis, and enhances simulation efficiency and accuracy.

[0086] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0087] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A multi-objective collaborative simulation and optimization system for ship structures, characterized in that, The system includes: The ship structure model building module is used to perform structural geometric parameterization based on the partition characteristics of the ship's physical structure and build a ship structure model. The simulation optimization target acquisition module is used to analyze the coupling relationship between expected functions, performance indicators and application scenarios, and generate multiple simulation optimization targets, which correspond to the expected functions, performance indicators and application scenarios respectively. The model configuration module is used to configure the computational model and the finite element analysis model according to the simulation optimization target; The simulation space construction module is used to couple the ship structure model, finite element analysis model, and computation model into a unified simulation space through a parameter mapping interface. The simulation test data output module is used to perform adaptive step size adjustment of the parameters of the finite element analysis model within the parameter range of the expected function, performance index and application scenario, and output simulation test data based on the adjusted parameters. An anomaly parameter acquisition module is used to perform calculations on the simulation test data through the calculation model to obtain test calculation results, and to identify anomalies in the test calculation results based on the simulation optimization target to obtain anomaly parameters; The optimization variable determination module is used to analyze and locate the abnormal parameters, determine the optimization variables, and optimize the optimization variables through a preset optimization algorithm to obtain ship structure optimization information.

2. The multi-objective collaborative simulation optimization system for ship structures as described in claim 1, characterized in that, The model configuration module is also used for: Based on the expected functions, performance indicators, and application scenarios, the parameters are parsed, and the correlation analysis is performed on the parsed parameters to construct a parameter topology structure, which includes parameter relationships and relationship coefficients. Based on the expected functions, performance indicators, and application scenarios, target evaluation parameters are set. Using the target evaluation parameters as the search direction, relationship extraction is performed based on the parameter topology to construct a target evaluation parameter relationship set; Training data is constructed based on the target evaluation parameter relationship set, and the model is trained using the training data until the target is converged to obtain the computational model, which has target evaluation parameter identification information.

3. The multi-objective collaborative simulation optimization system for ship structures as described in claim 1, characterized in that, The model configuration module is also used for: Based on the ship's physical structure, a coarse-grained grid is set. The coarse-grained grid is used to determine the grid division rules based on the ship's hull functional partitions, unit component partitions, and shape structure partitions of the ship's physical structure. Based on the simulation optimization objective, the target hull structure is determined, wherein the target hull structure is the hull structure that has an impact on the simulation optimization objective; An impact analysis is performed on the target hull structure, and a fine-grained mesh is set based on the impact. The size of the fine-grained mesh is inversely proportional to the impact; the greater the impact, the smaller the mesh. Define the properties of ship materials, including elastic modulus, thermal conductivity, density, and hardness; A finite element analysis model is constructed based on the coarse-grained mesh, fine-grained mesh, and ship material properties. The loads and boundary conditions are determined according to the simulation optimization objectives, and the loads and boundary conditions are fitted to the finite element analysis model.

4. The multi-objective collaborative simulation optimization system for ship structures as described in claim 3, characterized in that, The simulation test data output module is also used for: Based on the parameter range, set the test adjustment step size; Based on the test adjustment step size and parameter range, the parameters are adjusted, and the stress change is decomposed according to the coarse-grained grid and fine-grained grid to obtain the stress value change density of each grid. When the stress value change density reaches a preset threshold, the mesh parameters are adjusted based on the stress value gradient to determine the mesh configuration parameters. Based on the parameter range and test adjustment step size, the load parameters are adjusted to determine the load configuration parameters. The parameters of the finite element analysis model are then configured using the mesh configuration parameters and load configuration parameters.

5. The multi-objective collaborative simulation optimization system for ship structures as described in claim 1, characterized in that, The abnormal parameter acquisition module is also used for: Based on the aforementioned simulation optimization objectives, an abnormal case record dataset is obtained; Anomaly decomposition is performed based on the dataset of abnormal case records to determine the abnormal factors and corresponding abnormal thresholds. The abnormal factors are those that affect the abnormal case events. Based on the aforementioned abnormal factors and abnormal thresholds, an abnormal identification module is configured to identify abnormal factors in the test operation results and to determine abnormalities through the abnormal thresholds to obtain abnormal parameters, wherein the abnormal parameters are abnormal factors that exceed the abnormal thresholds.

6. The multi-objective collaborative simulation optimization system for ship structures as described in claim 5, characterized in that, The optimization variable determination module is also used for: Based on the aforementioned abnormal case record dataset, abnormal time series data decomposition is performed to construct time series training data; The time series prediction model is trained using the aforementioned time series training data to obtain an anomaly prediction model; The abnormal parameters are introduced into the anomaly prediction model. Starting from the result of the abnormal event risk, reverse parameter tracking is performed based on the anomaly prediction model, and the impact of parameter changes on the output result is quantified. The optimization variables are selected based on their impact, and the optimization variables are ship structural parameters whose impact is greater than a threshold.

7. The multi-objective collaborative simulation optimization system for ship structures as described in claim 6, characterized in that, The abnormal parameter acquisition module is also used for: When the abnormal parameter output is empty, an abnormal prediction is performed based on the test calculation results using the abnormal prediction model to obtain the abnormal prediction probability. Based on the anomaly prediction probability, risk events are screened, and the risk events are anomaly events whose anomaly prediction probability reaches a preset requirement; Based on the risk event and the test calculation results, the abnormal parameter is determined. The abnormal parameter is the abnormal factor corresponding to the risk event, and the difference between the abnormal factor and the abnormal threshold in the test calculation results is less than a preset condition.

8. The multi-objective collaborative simulation optimization system for ship structures as described in claim 6, characterized in that, The optimization variable determination module is also used for: Based on the optimization variables, a random population is established, and the random population is divided into a first population and a second population. Based on the first population and the second population, determine the optimization center and optimization direction; Based on the optimization variables, an analysis of the correlation and impact on the ship's physical structure is conducted to determine the correlation structural parameters and the linkage effect. Based on the associated structural parameters and the linkage influence, determine the constraints of the optimization variables; Based on the constraints, a search is performed using the optimization center and optimization direction, the search results are evaluated using a fitness function, and the population is updated based on the fitness. When the update frequency does not meet the update iteration requirements, the optimal solution of the current population is added to the taboo space, and the optimization is performed iteratively based on the neighborhood space of the current population. The update frequency judgment result is used as the taboo condition. When the optimization target or number of iterations is reached, the current optimal solution and the optimal solution in the taboo space are obtained to evaluate fitness and determine the ship structure optimization information.

9. A multi-objective collaborative simulation optimization method for ship structures, characterized in that, The method includes: Based on the partitioning characteristics of the ship's physical structure, structural geometric parameters are performed to construct a ship structural model. The coupling relationship between expected functions, performance indicators and application scenarios is analyzed to generate multiple simulation optimization objectives, which correspond to the objectives of expected functions, performance indicators and application scenarios respectively. Configure the computational model and finite element analysis model according to the simulation optimization objectives; The ship structure model, finite element analysis model, and computational model are coupled into a unified simulation space through a parameter mapping interface; Within the parameter range of the expected function, performance indicators and application scenarios, the adaptive step size adjustment of the finite element analysis model parameters is performed, and simulation test data is output based on the adjusted parameters. The simulation test data is processed by the computational model to obtain test results. Anomalies are identified in the test results based on the simulation optimization objective to obtain abnormal parameters. Based on the abnormal parameters, the location is analyzed and determined, the optimization variables are identified, and the optimization variables are optimized using a preset optimization algorithm to obtain ship structure optimization information.

Images