Intelligent optimization design method and system based on multi-dimensional sound and vibration coupling simulation of ship cabin

Abstract

The invention relates to an intelligent optimization design method and system based on multi-dimensional sound and vibration coupling simulation of a ship cabin. The method comprises the following steps: obtaining structural data of a ship cabin to be optimized and sound vibration data under multiple working conditions, and constructing an initial simulation model by adopting a finite element analysis method; simulating and acquiring sound vibration energy propagation paths and attenuation characteristic data under multiple working conditions based on the model, constructing a sound vibration coupling relation matrix, and identifying a key sound vibration parameter set; and classifying the key sound and vibration parameter set into a first-level parameter subset and a second-level parameter subset according to a preset screening rule, generating a preliminary ship cabin design scheme through a preset neural network model in combination with the initial simulation model and the first-level parameter subset, and performing multi-target collaborative optimization based on the second-level parameter subset to obtain an optimized design scheme. According to the method, the sound-vibration coupling rule can be excavated, the scientificity and optimization efficiency of ship cabin design are improved, and the sound-vibration performance of the cabin is effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of marine engineering technology, and in particular relates to an intelligent optimization design method and system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments. Background Technology

[0002] With the development of marine engineering technology, especially the continuous improvement of the requirements for ship comfort, safety and comprehensive performance, the optimization of the acoustic and vibration performance of ship cabins has become one of the core technical requirements in the field of ship design. Acoustic and vibration simulation technology has been gradually applied to the design process of ship cabins. It has the characteristic of initially simulating the acoustic and vibration response of cabin structure under specific working conditions, which can assist designers in predicting the acoustic and vibration performance of cabins to a certain extent.

[0003] Traditional technologies for ship cabin acoustic and vibration design often rely on empirical formulas or simplified single-condition simulation analyses. Specifically, this involves comparing parameters with previous similar ship cabin designs, or constructing simplified acoustic and vibration models for a single typical condition such as uniform ship navigation. Structural parameters such as bulkhead thickness, sound insulation material selection, and placement are manually adjusted to attempt to improve acoustic and vibration performance. The design effectiveness is then verified through physical prototype fabrication and actual ship testing. However, current methods have significant shortcomings: they fail to fully consider the complex propagation and coupling characteristics of acoustic and vibration energy under various conditions, including ship startup, acceleration, full load, and empty load, resulting in poor acoustic and vibration performance stability in real-world scenarios. Relying on experience and simplified single-condition models makes it difficult to accurately capture the propagation path and attenuation patterns of acoustic and vibration energy, leading to low design accuracy. Furthermore, manual parameter adjustment optimization is inefficient, unable to achieve multi-parameter collaborative optimization, and struggles to balance cabin acoustic and vibration performance, structural strength, and manufacturing costs, ultimately affecting the overall design quality and user experience of the ship cabin. Summary of the Invention

[0004] Therefore, it is necessary to provide an intelligent optimization design method and system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments that can solve the above problems.

[0005] Firstly, this application provides an intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments, including:

[0006] Acquire structural data and acoustic and vibration data under multiple operating conditions for the ship's compartments to be optimized;

[0007] Based on the structural data and acoustic and vibration data, an initial simulation model is constructed using the finite element analysis method;

[0008] Based on the initial simulation model, data on the propagation path and attenuation characteristics of acoustic and vibration energy under multiple operating conditions were obtained through simulation.

[0009] Based on the data of acoustic and vibration energy propagation paths and attenuation characteristics, an acoustic and vibration coupling relationship matrix of the initial simulation model is constructed.

[0010] Based on the acoustic-vibration coupling matrix, the key acoustic-vibration parameter set under multiple working conditions is identified.

[0011] The key acoustic and vibration parameter sets are classified and filtered according to the preset screening rules to form a graded acoustic and vibration parameter set, which includes a primary parameter subset and a secondary parameter subset.

[0012] By combining the initial simulation model with a subset of first-level parameters, a preliminary ship compartment design scheme is generated through a pre-set neural network model.

[0013] Based on a subset of secondary parameters, a multi-objective collaborative optimization of the preliminary ship compartment design scheme is performed to obtain an optimized ship compartment design scheme.

[0014] In one embodiment, based on an initial simulation model, data on the propagation path and attenuation characteristics of acoustic and vibration energy under multiple operating conditions are simulated and obtained, including:

[0015] Based on acoustic and vibration data under multiple working conditions, the structural constraints and acoustic and vibration excitation conditions corresponding to each working condition are determined in the initial simulation model.

[0016] Based on structural constraints and acoustic vibration excitation conditions, an initial simulation model is used to perform acoustic vibration response simulation calculations, and the acoustic vibration response simulation results are obtained.

[0017] From the acoustic vibration response simulation results, the acoustic vibration characteristic locations where acoustic vibration energy converges and is transferred are selected;

[0018] Based on the acoustic vibration characteristic locations, the temporal variation data of acoustic vibration energy at each characteristic location in the acoustic vibration response simulation results are extracted;

[0019] The structural connection method and medium energy transfer law of the acoustic vibration characteristic location are obtained, and the acoustic vibration characteristic location, structural connection method and medium energy transfer law are integrated to form energy propagation nodes;

[0020] By combining time-series variation data with energy propagation nodes, the transmission trajectory of acoustic and vibration energy from the excitation source to the ship's compartments is traced in reverse to obtain acoustic and vibration energy propagation path data;

[0021] For each path in the acoustic vibration energy propagation path data, the energy transfer is divided into different stages;

[0022] Based on the acoustic vibration response simulation results, the energy amplitude attenuation and attenuation rate of the acoustic vibration energy propagation path in each stage are calculated to obtain attenuation characteristic data.

[0023] In one embodiment, based on the acoustic energy propagation path and attenuation characteristic data, an acoustic coupling matrix of the initial simulation model is constructed, including:

[0024] The number of rows and columns of the matrix to be constructed is determined based on the energy propagation nodes in the acoustic energy propagation path data.

[0025] Based on the energy propagation nodes and the number of rows and columns in the matrix, establish a unique correspondence between each energy propagation node and the matrix row and column indices;

[0026] Traverse the acoustic energy propagation path data and, based on the unique correspondence, record the matrix row and column index combination of adjacent energy propagation nodes on each path;

[0027] For each combination of matrix row and column indices, the acoustic-vibration coupling coefficient is calculated based on the attenuation characteristic data;

[0028] Fill the acoustic-vibration coupling coefficients into the cells of the corresponding row and column indices in the matrix to be constructed, and assign preset initial values ​​to the cells of the matrix to be constructed that have not been filled with data, thereby generating the acoustic-vibration coupling relationship matrix.

[0029] In one embodiment, the acoustic-vibration coupling matrix is ​​calculated using the following formula:

[0030]

[0031] in, C is the acoustic-vibration coupling matrix, and n is the total number of nodes for acoustic-vibration energy propagation. ij E represents the acoustic-vibration energy coupling coefficient from node i to node j. trans，ij E represents the acoustic energy transmitted from node i to node j. soure，i d represents the initial acoustic energy value of node i. ij β represents the effective propagation distance between node i and node j, and β represents the energy attenuation coefficient.

[0032] In one embodiment, based on the acoustic-vibration coupling matrix, a set of key acoustic-vibration parameters under multiple operating conditions is identified, including:

[0033] For each working condition in the multi-condition scenario, the coupling influence weight of each acoustic and vibration parameter under the corresponding working condition is calculated based on the acoustic-vibration coupling relationship matrix.

[0034] The coupling influence weights are compared with preset weight thresholds, and acoustic vibration parameters with coupling influence weights greater than preset weight thresholds are selected to form a subset of candidate acoustic vibration parameters for each working condition.

[0035] Within the subset of candidate acoustic and vibration parameters corresponding to each working condition, the cross-working condition coverage ratio of each acoustic and vibration parameter is statistically analyzed.

[0036] The cross-condition coverage ratio is compared with a preset ratio threshold, and acoustic and vibration parameters with a cross-condition coverage ratio greater than the preset ratio threshold are selected to form a set of key acoustic and vibration parameters under multiple conditions.

[0037] In one embodiment, the preset screening rule includes a first-level parameter judgment condition and a second-level parameter judgment condition; wherein, the first-level parameter judgment condition is that the coupling influence weight is greater than or equal to the preset first-level weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset first-level coverage threshold; the second-level parameter judgment condition is that the coupling influence weight is greater than or equal to the preset second-level weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset second-level coverage threshold; the preset first-level weight threshold is greater than the preset second-level weight threshold, and the preset first-level coverage threshold is greater than the preset second-level coverage threshold;

[0038] The key acoustic and vibration parameter sets are classified and filtered according to preset screening rules to form a graded acoustic and vibration parameter set, including:

[0039] Determine whether each key acoustic and vibration parameter meets the first-level parameter judgment condition, generate the first-level judgment result, and based on the first-level judgment result, classify the key acoustic and vibration parameters that meet the first-level parameter judgment condition into the first-level parameter subset, and classify the key acoustic and vibration parameters that do not meet the first-level parameter judgment condition into the second-level candidate parameter subset.

[0040] Determine whether the key acoustic and vibration parameters in the secondary candidate parameter subset meet the secondary parameter judgment conditions, generate secondary judgment results, and classify the key acoustic and vibration parameters that meet the secondary parameter judgment conditions into the secondary parameter subset based on the secondary judgment results;

[0041] The primary and secondary parameter subsets are integrated to form a hierarchical acoustic and vibration parameter set.

[0042] In one embodiment, based on a subset of secondary parameters, a multi-objective collaborative optimization is performed on the preliminary ship compartment design scheme to obtain an optimized ship compartment design scheme, including:

[0043] Based on the key acoustic and vibration parameters in the secondary parameter subset, the optimization indicators of the scheme are determined;

[0044] Based on the scheme optimization indicators, and combined with the preset ship compartment design specifications and manufacturing requirements, the optimization objectives and constraints of multi-objective collaborative optimization are set.

[0045] Based on the optimization objectives and constraints, the preliminary ship cabin design scheme is adjusted using a pre-set neural network model to form an optimized ship cabin design scheme.

[0046] Secondly, this application also provides an intelligent optimization design system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments, including:

[0047] The compartment data acquisition module is used to acquire structural data and acoustic and vibration data of the compartments of the ship to be optimized under multiple operating conditions;

[0048] The simulation model building module is used to build an initial simulation model based on structural data and acoustic vibration data using the finite element analysis method.

[0049] The acoustic and vibration characteristics simulation module is used to simulate and obtain acoustic and vibration energy propagation paths and attenuation characteristics data under multiple operating conditions based on the initial simulation model.

[0050] The relation matrix construction module is used to construct the acoustic-vibration coupling relation matrix of the initial simulation model based on the acoustic-vibration energy propagation path and attenuation characteristic data.

[0051] The key parameter identification module is used to identify the set of key acoustic and vibration parameters under multiple working conditions based on the acoustic-vibration coupling relationship matrix.

[0052] The parameter classification and filtering module is used to classify and filter the key acoustic and vibration parameter sets according to preset filtering rules to form a graded acoustic and vibration parameter set, which includes a primary parameter subset and a secondary parameter subset.

[0053] The design scheme generation module is used to combine the initial simulation model with a subset of first-level parameters and generate preliminary ship compartment design schemes through a preset neural network model.

[0054] The scheme collaborative optimization module is used to perform multi-objective collaborative optimization on the preliminary ship compartment design scheme based on a subset of secondary parameters, so as to obtain an optimized ship compartment design scheme.

[0055] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-mentioned intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship cabins;

[0056] Fourthly, this application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-mentioned intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship cabins.

[0057] The aforementioned intelligent optimization design method and system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments acquires structural data and acoustic-vibration data of the ship compartment to be optimized under multiple operating conditions. It then uses finite element analysis to construct an initial simulation model, simulating the acoustic-vibration energy propagation path and attenuation characteristics under multiple operating conditions and constructing an acoustic-vibration coupling matrix. It identifies key acoustic-vibration parameter sets under multiple operating conditions and classifies them into primary and secondary parameter subsets. Combining the initial simulation model and the primary parameter subsets, it generates a preliminary design scheme through a pre-set neural network model. Based on the secondary parameter subsets, it performs multi-objective collaborative optimization, effectively uncovering the acoustic-vibration coupling laws of ship compartments under multiple operating conditions. This solves the problems of traditional design methods lacking in-depth consideration of acoustic-vibration coupling relationships, insufficient design scientificity, low optimization efficiency, and poor cabin acoustic-vibration performance. It improves the scientificity and optimization efficiency of ship compartment design and enhances cabin acoustic-vibration performance. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart of the intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments according to the present invention;

[0060] Figure 2 This is a structural diagram of the intelligent optimization design system based on multi-dimensional acoustic-vibration coupling simulation of ship cabins according to the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] In one embodiment, such as Figure 1 As shown, an intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship cabins is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, or to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In the implementation environment, the hardware devices include acoustic-vibration sensors, structural scanning sensors, a terminal, and a server. The application scenario includes: when the design of a ship cabin requires improvement in acoustic-vibration performance, the terminal collects cabin structural data and multi-condition acoustic-vibration data through sensors, transmits them to the server, the server uses finite element analysis to construct a model, simulates acoustic-vibration characteristics, and constructs a coupling matrix, generates a preliminary scheme using a neural network, optimizes based on secondary parameters, and then sends the optimized scheme back to the terminal, completing the interaction. In this embodiment, the method includes the following steps:

[0063] S01, acquire structural data of the ship's compartments to be optimized and acoustic and vibration data under multiple operating conditions.

[0064] Among them, the ship compartments to be optimized refer to various functional compartments inside the ship (including but not limited to living quarters, equipment compartments, control rooms, etc.) whose acoustic and vibration performance needs to be improved. The structural data covers basic data related to the compartment structure, such as compartment geometry, bulkhead material properties, component connection methods, and internal equipment layout parameters. The multi-operating conditions include typical operating states such as ship start-up, acceleration, constant speed, deceleration, full load, and empty load, as well as special operating conditions that can be included according to design requirements. The acoustic and vibration data refer to relevant data reflecting the acoustic and vibration response, such as vibration acceleration, sound pressure level, vibration frequency, and sound wave propagation characteristics generated by the compartment structure and internal environment under multiple operating conditions. In practice, the structural data can be determined by 3D scanning of existing compartments, extracting from original design drawings and technical documents, or based on the initial structural scheme preset based on the design objectives. The acoustic and vibration data can be collected by placing acoustic and vibration sensors (such as acceleration sensors, microphones, etc.) at key locations in the compartment under simulated or actual multi-operating conditions, or obtained by retrieving from a multi-operating condition acoustic and vibration database of similar ships and combining it with the characteristics of the compartment to be optimized.

[0065] S02. Based on the structural data and acoustic vibration data, an initial simulation model is constructed using the finite element analysis method.

[0066] The finite element method (FEM) involves discretizing the ship's cabin structure to be optimized into a finite number of interconnected structural and acoustic elements. It then establishes element mechanical equations and assembles them into an overall system equation to solve the acoustic-vibration coupling problem. The initial simulation model is a digital model capable of initially simulating the acoustic-vibration characteristics of the ship's cabin under various operating conditions. In implementation, the structural and acoustic-vibration data can be imported into finite element software with acoustic-vibration analysis capabilities (including but not limited to ANSYS, ABAQUS, etc.). Based on the cabin structure's complexity and analysis accuracy requirements, the cabin structure is meshed using finite element methods (mesh size adapted to the acoustic-vibration analysis frequency range to ensure computational effectiveness). Corresponding material properties are assigned to each mesh element, and component connection constraints are set. Combined with the excitation information and response reference data from the acoustic-vibration data, load boundary conditions (such as excitation loading position and direction) and acoustic boundary conditions (such as cabin wall sound absorption characteristics) are defined in the model. The model is then initialized and verified using finite element software (eliminating mesh quality defects, incorrect boundary condition settings, etc.), resulting in an initial simulation model that can be used for subsequent acoustic-vibration characteristic simulation.

[0067] S03, based on the initial simulation model, simulate and obtain data on the propagation path and attenuation characteristics of acoustic and vibration energy under multiple working conditions.

[0068] Among them, the acoustic and vibration energy propagation path refers to the trajectory of acoustic and vibration energy from external excitation sources (such as hull vibration, propulsion system vibration) or internal excitation sources (such as equipment vibration inside the cabin) to various areas of the cabin through the cabin structural components (such as bulkheads, decks) and internal media (such as air, liquid); the attenuation characteristic data refers to the amount of energy amplitude reduction, attenuation rate and attenuation differences at different propagation stages caused by factors such as medium loss, structural damping, and interface reflection during the propagation process. In implementation, based on the initial simulation model and multi-condition acoustic and vibration data, corresponding structural constraints and acoustic and vibration excitation conditions are configured for each condition in the finite element analysis software. The full-field data of the acoustic and vibration response of the compartment under each condition are obtained through the software solver. Using the post-processing function of the simulation software (or a custom energy tracking algorithm), the key feature locations of acoustic and vibration energy convergence and transmission are identified from the response data. The transmission order and correlation of energy between feature locations are tracked to determine the acoustic and vibration energy propagation path under multiple conditions. The energy value changes of different segments on each propagation path are extracted, and the energy attenuation amount and attenuation rate of the corresponding segment are calculated. These are integrated to form acoustic and vibration energy attenuation characteristic data under multiple conditions, providing effective support for the subsequent construction of the acoustic and vibration coupling relationship matrix.

[0069] S04. Based on the acoustic and vibration energy propagation path and attenuation characteristic data, construct the acoustic and vibration coupling relationship matrix of the initial simulation model.

[0070] The acoustic-vibration coupling matrix is ​​used to quantitatively describe the energy coupling strength and transmission relationship between each acoustic-vibration energy propagation node in the initial simulation model. In implementation, key acoustic-vibration energy propagation nodes (such as structural connection points where energy converges, medium boundary points, etc.) can be extracted from the acoustic-vibration energy propagation path. The number of rows and columns of the acoustic-vibration coupling matrix is ​​determined based on the number of nodes, establishing a unique correspondence between each propagation node and the matrix row and column indices. The acoustic-vibration energy propagation path data is traversed, and combined with attenuation characteristic data, the acoustic-vibration coupling coefficient between each pair of adjacent propagation nodes is determined through methods such as energy transfer ratio calculation and attenuation coefficient correction (e.g., based on the ratio of energy transfer between nodes to the initial energy of the node, combined with the attenuation effect corresponding to the propagation distance). The calculated coupling coefficients are then filled into the cells of the corresponding row and column indices in the matrix. Preset initial values ​​are assigned to cells in the matrix that are not involved in energy transfer, thus completing the construction of the acoustic-vibration coupling matrix of the initial simulation model.

[0071] S05, based on the acoustic-vibration coupling relationship matrix, identifies the key acoustic-vibration parameter set under multiple working conditions.

[0072] The key acoustic and vibration parameter set refers to the set of acoustic and vibration-related parameters (including but not limited to bulkhead material damping coefficient, component connection stiffness, internal acoustic material sound absorption coefficient, excitation source characteristic parameters, etc.) that have a significant impact on the acoustic and vibration performance (such as sound pressure level and vibration acceleration) of the ship's cabins to be optimized under multiple operating conditions and require key control. In implementation, based on the coupling coefficients between nodes in the acoustic and vibration coupling relationship matrix, quantitative analysis methods (such as parameter sensitivity analysis, coupling contribution calculation, etc., which can be implemented through the built-in functions of finite element analysis software or custom algorithms) can be used to calculate the coupling influence weight of each acoustic and vibration parameter on the cabin's acoustic and vibration response under each operating condition. A preset weight threshold is set, and acoustic and vibration parameters with coupling influence weights greater than this threshold under each operating condition are selected to form a subset of candidate acoustic and vibration parameters for the corresponding operating condition. The frequency of occurrence of each candidate acoustic and vibration parameter in all candidate subsets under all operating conditions is counted, and its cross-operating condition coverage ratio is calculated. A preset ratio threshold is set, and acoustic and vibration parameters with cross-operating condition coverage ratios greater than this threshold are selected and integrated to form the key acoustic and vibration parameter set under multiple operating conditions.

[0073] S06. According to the preset screening rules, the key acoustic and vibration parameter set is classified and screened to form a graded acoustic and vibration parameter set, which includes a primary parameter subset and a secondary parameter subset.

[0074] The preset screening rules are parameter classification judgment standards pre-set based on the optimization goals of the acoustic and vibration performance of the ship's cabin to be optimized (such as noise reduction and vibration reduction indicators), engineering design accuracy requirements, and industry practice experience. These standards include core evaluation indicators and corresponding thresholds used to distinguish the priority of parameter impact. The graded acoustic and vibration parameter set refers to the parameter set formed by dividing key acoustic and vibration parameters according to their degree of impact on acoustic and vibration performance and control priority. The first-level parameter subset is the set of parameters that play a core leading role in the acoustic and vibration performance of the cabin, can significantly improve acoustic and vibration performance after optimization and adjustment, and should be given priority in design control. The second-level parameter subset is the set of parameters that have an important impact on the acoustic and vibration performance of the cabin but have a lower control priority than the first-level parameters and need to be further adjusted based on the optimization of the first-level parameters. During implementation, preset filtering rules are invoked. Each parameter in the key acoustic and vibration parameter set is compared with the first-level parameter judgment conditions according to the judgment conditions corresponding to the first-level and second-level parameters. Parameters that meet the first-level judgment conditions are assigned to the first-level parameter subset, while parameters that do not meet the first-level judgment conditions are further compared with the second-level parameter judgment conditions. Parameters that meet the second-level judgment conditions are assigned to the second-level parameter subset. The first-level parameter subset and the second-level parameter subset are then integrated to form a graded acoustic and vibration parameter set.

[0075] S07, combining the initial simulation model with a subset of first-level parameters, generates a preliminary ship cabin design scheme through a preset neural network model.

[0076] The preset neural network model is an artificial intelligence model (such as a BP neural network or convolutional neural network) that is pre-trained based on historical design data, acoustic and vibration simulation data, and actual ship verification data in the field of ship cabin design. It has the ability to learn the mapping relationship between design parameters and acoustic and vibration performance. During implementation, the structural feature data of the initial simulation model (such as the overall geometric topology of the cabin and the initial parameters of the core components) and the parameters of the first-level parameter subset are input into the preset neural network model. Based on the trained mapping relationship, the model performs inference calculations in combination with the acoustic and vibration performance optimization objectives (such as reducing the sound pressure level inside the cabin and controlling the vibration acceleration) to generate multiple sets of candidate design schemes that meet the constraints of the first-level parameters. The candidate schemes are preliminarily verified for compliance (such as checking whether they meet the strength specifications of the basic structure of the ship cabin and whether the first-level parameters are within a reasonable range), and the schemes that meet the basic requirements are selected as the preliminary ship cabin design schemes.

[0077] S08, based on a subset of secondary parameters, performs multi-objective collaborative optimization on the preliminary ship compartment design scheme to obtain an optimized ship compartment design scheme.

[0078] Among them, multi-objective collaborative optimization refers to the process of comprehensively considering multiple design objectives such as the acoustic and vibration performance of ship cabins (such as cabin sound pressure level and structural vibration acceleration), structural strength (such as load-bearing capacity), and manufacturing cost (such as material usage and processing difficulty), and achieving a balance between the objectives by adjusting parameters; optimizing ship cabin design schemes refers to digital design schemes that meet the preset acoustic and vibration performance indicators, structural safety specifications, and cost control requirements after multi-objective collaborative optimization. In implementation, based on the characteristics of each parameter in the secondary parameter subset and the existing state of the preliminary ship compartment design scheme, combined with the preset ship compartment design specifications and manufacturing requirements, the specific objectives of multi-objective collaborative optimization (such as reducing the sound pressure level to XX decibels and controlling the manufacturing cost within XX range) and constraints (such as the secondary parameter values ​​not exceeding the material performance limits) are determined. Algorithms suitable for multi-objective optimization (such as genetic algorithms, particle swarm optimization algorithms, etc., which can be implemented in conjunction with computer simulation tools) are used, with the secondary parameters as adjustment variables, to iteratively optimize the corresponding parameters in the preliminary ship compartment design scheme. The achievement of each optimization objective is verified in real time to ensure a synergistic balance among the objectives. The optimized scheme undergoes comprehensive performance verification (such as recalculating acoustic and vibration performance through the initial simulation model and checking structural strength and cost indicators). After confirming that all preset requirements are met, the optimized ship compartment design scheme is obtained.

[0079] The aforementioned intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments first acquires structural data and acoustic-vibration data of the ship compartment to be optimized under multiple operating conditions. It then uses finite element analysis to construct an initial simulation model, overcoming the limitations of traditional simplified single-condition models and achieving coverage of multiple operating conditions. Based on this initial simulation model, it simulates and acquires data on acoustic-vibration energy propagation paths and attenuation characteristics under multiple operating conditions, constructs an acoustic-vibration coupling relationship matrix, and identifies key acoustic-vibration parameter sets, solving the problem that traditional methods struggle to accurately capture acoustic-vibration coupling patterns. According to preset screening rules, the key acoustic-vibration parameter sets are classified into primary and secondary parameter subsets. Combining the initial simulation model and the primary parameter subsets, a preliminary design scheme is generated through a preset neural network model. Multi-objective collaborative optimization is then performed based on the secondary parameter subsets, replacing the inefficient traditional manual parameter adjustment method and achieving multi-parameter collaborative control.

[0080] In one embodiment, based on an initial simulation model, data on the propagation path and attenuation characteristics of acoustic and vibration energy under multiple operating conditions are simulated and obtained, including:

[0081] S11, Based on acoustic and vibration data under multiple working conditions, determine the structural constraints and acoustic and vibration excitation conditions corresponding to each working condition in the initial simulation model;

[0082] S12, Based on the structural constraints and acoustic vibration excitation conditions, the initial simulation model is used to perform acoustic vibration response simulation calculations to obtain the acoustic vibration response simulation results;

[0083] S13, From the acoustic vibration response simulation results, select the acoustic vibration characteristic locations where acoustic vibration energy converges and is transferred;

[0084] S14, Based on the acoustic vibration characteristic locations, extract the time-series variation data of acoustic vibration energy at each characteristic location in the acoustic vibration response simulation results;

[0085] S15, obtain the structural connection method and medium energy transfer law of the acoustic vibration characteristic location, and integrate the acoustic vibration characteristic location, structural connection method and medium energy transfer law to form an energy propagation node;

[0086] S16, combining time-series change data and energy propagation nodes, reverse the transmission trajectory of acoustic and vibration energy from the excitation source to the ship's compartments to obtain acoustic and vibration energy propagation path data;

[0087] S17, for each path in the acoustic vibration energy propagation path data, divide the energy transfer into different stages;

[0088] S18. Based on the acoustic vibration response simulation results, calculate the energy amplitude attenuation and attenuation rate of the acoustic vibration energy propagation path in each stage to obtain attenuation characteristic data.

[0089] For example, based on acoustic and vibration data under multiple operating conditions, the structural constraints (such as component connection and fixing methods, boundary support status) and acoustic and vibration excitation conditions (such as excitation loading location, direction, and energy amplitude) can be clearly defined in the initial simulation model in conjunction with the ship's operating state (such as start-up, acceleration, full load, and empty load) for each operating condition. Based on these constraints and excitation conditions, the initial simulation model can be called using finite element analysis software to perform acoustic and vibration response simulation calculations, obtaining acoustic and vibration response simulation results covering vibration acceleration, sound pressure level, and energy distribution in various areas of the cabin. From these simulation results, the acoustic and vibration characteristic locations where acoustic and vibration energy is concentrated and mutually transmitted can be screened out through the finite element software post-processing function or a custom energy convergence identification algorithm. Based on these acoustic and vibration characteristic locations, the time-series variation data of acoustic and vibration energy at each characteristic location in the simulation results can be extracted. At the same time, the structural connection methods (such as welding, bolt connection, etc.) at each acoustic and vibration characteristic location can be obtained by retrieving the cabin structural design file or using structural scanning, and combined with acoustics and mechanics The principle determines the energy transfer law of the medium at the corresponding location (such as the sound wave transmission law in the air medium and the vibration energy transfer law in the structural components). The sound and vibration characteristic locations, structural connection methods and medium energy transfer laws are integrated to form energy propagation nodes that can characterize key energy transfer nodes. Combining the extracted time-series change data with the constructed energy propagation nodes, the energy is traced backward in the order of strong to weak and back to front to determine the complete trajectory of sound and vibration energy from the excitation source (such as hull vibration and in-cabin equipment vibration) to various areas of the ship's cabin, and the sound and vibration energy propagation path data is obtained. For each propagation path, according to the characteristics such as changes in medium type, distribution of structural connection nodes or abrupt changes in energy transfer law, different stages of energy transfer are divided. Based on the energy values ​​corresponding to each stage in the sound and vibration response simulation results, the energy amplitude attenuation is obtained by calculating the energy difference between adjacent characteristic locations. The attenuation rate is calculated by the ratio of the attenuation to the corresponding propagation distance or time, and attenuation characteristic data covering the attenuation differences of each stage is obtained.

[0090] In one embodiment, based on the acoustic energy propagation path and attenuation characteristic data, an acoustic coupling matrix of the initial simulation model is constructed, including:

[0091] S21, Determine the number of rows and columns of the matrix to be constructed based on the energy propagation nodes in the acoustic energy propagation path data;

[0092] S22, Based on the energy propagation nodes and the number of rows and columns of the matrix, establish a unique correspondence between each energy propagation node and the matrix row and column index;

[0093] S23, Traverse the acoustic energy propagation path data, and record the matrix row and column index combination of adjacent energy propagation nodes on each path, based on the unique correspondence.

[0094] S24, For each combination of matrix row and column indices, calculate the acoustic-vibration coupling coefficient based on the attenuation characteristic data;

[0095] S25, fill the acoustic-vibration coupling coefficient into the cell corresponding to the row and column index in the matrix to be constructed, and assign a preset initial value to the cell in the matrix to be constructed that has not been filled with data, and generate the acoustic-vibration coupling relationship matrix.

[0096] Specifically, the number of rows and columns of the matrix to be constructed can be determined based on the energy propagation nodes already integrated in the acoustic and vibration energy propagation path data. Since the acoustic and vibration coupling relationship matrix needs to represent the coupling relationship between all energy propagation nodes, the number of rows and columns of the matrix is ​​equal to the total number of energy propagation nodes, n. Based on the energy propagation nodes and the determined number of rows and columns of the matrix, a unique correspondence between each energy propagation node and the matrix row and column indices is established by assigning a unique serial number (e.g., node 1, node 2, ..., node n) to each energy propagation node. For example, node i corresponds to the index of the i-th row and the i-th column of the matrix. Each propagation path in the acoustic and vibration energy propagation path data is traversed, and adjacent energy propagation nodes in the energy transfer process are traced along each path. Combining the unique correspondence established above, the matrix row and column indices corresponding to adjacent nodes are recorded. Record the index combination (e.g., if node i and node j are adjacent, record it as (i,j)); for each set of recorded matrix row and column index combination (i,j), based on the acoustic and vibration energy transfer value from node i to node j, the initial acoustic and vibration energy value of node i, the effective propagation distance between node i and j, and the energy attenuation coefficient in the attenuation characteristic data, calculate the acoustic and vibration coupling coefficient corresponding to the set of indexes; fill the calculated acoustic and vibration coupling coefficients into the cells of the corresponding row and column indexes in the matrix to be constructed; for cells in the matrix to be constructed that do not record index combinations (i.e., there is no direct acoustic and vibration energy transfer relationship), assign a preset initial value (e.g., 0 or a very small value, which can be set according to the simulation accuracy requirements); after completing all coefficient filling and initial value assignment, generate the acoustic and vibration coupling relationship matrix of the initial simulation model.

[0097] In one embodiment, S31, the acoustic-vibration coupling matrix is ​​calculated using the following formula:

[0098]

[0099] in, C is the acoustic-vibration coupling matrix, and n is the total number of nodes for acoustic-vibration energy propagation. ij E represents the acoustic-vibration energy coupling coefficient from node i to node j. trans，ij E represents the acoustic energy transmitted from node i to node j. soure，i d represents the initial acoustic energy value of node i. ij β represents the effective propagation distance between node i and node j, and β represents the energy attenuation coefficient.

[0100] For example, first determine the dimension of the acoustic-vibration coupling relationship matrix C, which is an n-order square matrix, where n equals the total number of energy propagation nodes in the acoustic-vibration energy propagation path data, and each row and column of the matrix uniquely corresponds to an energy propagation node; any element C in the matrix ij (i.e., the element in the i-th row and j-th column) represents the acoustic-vibration energy coupling coefficient from energy propagation node i to node j, which is expressed by the formula... The calculation shows that E trans，ij The acoustic energy value transferred from node i to node j is obtained by extracting the energy transfer record between nodes i and j from the acoustic energy propagation path data; E soure，i d represents the initial acoustic vibration energy value of node i, which is derived from the energy data of node i at the initial excitation moment in the acoustic vibration response simulation results or the initial energy parameters of the excitation source applied to node i; ij β is the effective propagation distance between node i and node j, which is determined by measuring the spatial straight-line distance or the actual length of the energy transfer path between nodes i and j in the actual cabin structure using the structural data (such as a 3D structural model) of the ship's cabin to be optimized; β is the energy attenuation coefficient, which is obtained by referring to the acoustic and structural damping characteristics of the medium (such as air, bulkhead steel, sound insulation materials, etc.) through which energy transfer occurs between nodes i and j, referring to the acoustic vibration attenuation coefficient handbook of commonly used materials in the field of marine engineering, or by calibrating through acoustic vibration attenuation experiments of similar media; C is calculated for all node pairs with energy transfer relationships. ij Then, each C ij Fill in the cells corresponding to the row and column indices of the acoustic-vibration coupling matrix C. For cells that do not have a direct energy transfer relationship (i.e., node pairs not recorded in the acoustic-vibration energy propagation path data), assign a preset initial value (this initial value is usually set to 0 or a value much smaller than the effective coupling coefficient, which can be determined through pre-experiment verification according to the accuracy requirements of acoustic-vibration analysis of ship cabins) to assemble a complete acoustic-vibration coupling matrix C.

[0101] In one embodiment, based on the acoustic-vibration coupling matrix, a set of key acoustic-vibration parameters under multiple operating conditions is identified, including:

[0102] S41, For each working condition in the multi-working condition, calculate the coupling influence weight of each acoustic and vibration parameter under the corresponding working condition based on the acoustic-vibration coupling relationship matrix;

[0103] S42, compare the coupling influence weights with the preset weight thresholds respectively, and filter out the acoustic vibration parameters whose coupling influence weights are greater than the preset weight thresholds to form a subset of candidate acoustic vibration parameters for each working condition;

[0104] S43, within the subset of candidate acoustic and vibration parameters corresponding to each working condition, calculate the cross-working condition coverage ratio of each acoustic and vibration parameter;

[0105] S44 compares the cross-condition coverage ratio with the preset ratio threshold, and filters out the acoustic and vibration parameters whose cross-condition coverage ratio is greater than the preset ratio threshold, forming a set of key acoustic and vibration parameters under multiple conditions.

[0106] Specifically, for each of the ship's various operating conditions, such as starting, accelerating, constant speed, deceleration, full load, and empty load, the coupling coefficients between energy propagation nodes in the acoustic-vibration coupling relationship matrix can be analyzed using a parameter sensitivity analysis algorithm (such as the parameter sensitivity module built into finite element analysis software or a custom coupling contribution calculation program). This analysis examines the impact of minute changes in each acoustic-vibration parameter (such as the damping coefficient of bulkhead material, the stiffness of component connections, etc.) on the relevant coupling coefficients in the matrix. This impact is then normalized and converted into the coupling influence weight of each acoustic-vibration parameter under the corresponding operating condition. The calculated coupling influence weights of each acoustic-vibration parameter are then compared one by one with a preset weight threshold (this threshold is based on the ship's cabin acoustic-vibration performance optimization target, and is pre-set with reference to the performance compliance requirements in the ship cabin acoustic-vibration design specifications and key parameter screening cases of similar ships; for example, it is set to 0.6, but can be adjusted according to actual design accuracy requirements). Parameters with coupling influence weights greater than [a certain threshold] are then selected. The acoustic and vibration parameters with preset weight thresholds are categorized and integrated to form a subset of candidate acoustic and vibration parameters corresponding to each operating condition. Within the subsets of candidate acoustic and vibration parameters corresponding to all operating conditions, the number of times each acoustic and vibration parameter appears in different candidate subsets of operating conditions is counted. This number is divided by the total number of operating conditions (e.g., if there are 6 operating conditions and a certain acoustic and vibration parameter appears in the candidate subsets of 4 operating conditions, the count is 4) to obtain the cross-operating condition coverage ratio of each acoustic and vibration parameter (in the aforementioned example, the coverage ratio is 4 / 6≈66.7%). Finally, the cross-operating condition coverage ratio of each acoustic and vibration parameter is compared with a preset ratio threshold (this threshold is set to ensure that the parameter can have a significant impact on acoustic and vibration performance in most operating conditions, such as 70%, which can be adjusted according to the importance weight of the operating conditions). Acoustic and vibration parameters with a cross-operating condition coverage ratio greater than the preset ratio threshold are selected, and these parameters are integrated to form a set of key acoustic and vibration parameters under multiple operating conditions that can stably affect the acoustic and vibration performance of the cabin under multiple operating conditions.

[0107] In one embodiment, the preset screening rule includes a first-level parameter judgment condition and a second-level parameter judgment condition; wherein, the first-level parameter judgment condition is that the coupling influence weight is greater than or equal to the preset first-level weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset first-level coverage threshold; the second-level parameter judgment condition is that the coupling influence weight is greater than or equal to the preset second-level weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset second-level coverage threshold; the preset first-level weight threshold is greater than the preset second-level weight threshold, and the preset first-level coverage threshold is greater than the preset second-level coverage threshold;

[0108] The key acoustic and vibration parameter sets are classified and filtered according to preset screening rules to form a graded acoustic and vibration parameter set, including:

[0109] S51, determine whether each key acoustic and vibration parameter meets the first-level parameter judgment condition, generate the first-level judgment result, and according to the first-level judgment result, classify the key acoustic and vibration parameters that meet the first-level parameter judgment condition into the first-level parameter subset, and classify the key acoustic and vibration parameters that do not meet the first-level parameter judgment condition into the second-level candidate parameter subset.

[0110] S52, determine whether the key acoustic and vibration parameters in the secondary candidate parameter subset meet the secondary parameter judgment conditions, generate the secondary judgment results, and classify the key acoustic and vibration parameters that meet the secondary parameter judgment conditions into the secondary parameter subset according to the secondary judgment results;

[0111] S53 integrates the primary parameter subset and the secondary parameter subset to form a graded acoustic and vibration parameter set.

[0112] For example, in the preset screening rules, both the primary and secondary parameter judgment conditions are set based on the priority and control requirements of the impact of acoustic and vibration parameters on the acoustic and vibration performance of the cabin. The primary parameter judgment condition is that the coupling influence weight of the acoustic and vibration parameter is greater than or equal to the preset primary weight threshold, and its cross-condition coverage ratio is greater than or equal to the preset primary coverage threshold. The secondary parameter judgment condition is that the coupling influence weight of the acoustic and vibration parameter is greater than or equal to the preset secondary weight threshold, and its cross-condition coverage ratio is greater than or equal to the preset secondary coverage threshold. Simultaneously, the preset primary weight threshold (set based on the core acoustic and vibration performance optimization requirements of the cabin, referencing the key parameter control standards of similar ships, e.g., 0.8) is greater than the preset secondary weight threshold (set based on secondary acoustic and vibration performance optimization requirements, e.g., 0.5), and the preset primary coverage threshold (ensuring the parameter plays a dominant role in most operating conditions, e.g., 0.9) is greater than the preset secondary coverage threshold (ensuring the parameter plays an important role in most operating conditions, e.g., 0.6). In implementation, for each parameter in the key acoustic and vibration parameter set, the parameters are retrieved one by one. The coupling influence weight and cross-condition coverage ratio of the parameters under multiple operating conditions are compared with the preset first-level weight threshold and preset first-level coverage threshold in the first-level parameter judgment conditions to determine whether the parameter simultaneously meets the two threshold requirements, generating a first-level judgment result. Based on this result, all key acoustic and vibration parameters that meet the first-level judgment conditions are uniformly classified into the first-level parameter subset, while key acoustic and vibration parameters that do not meet the first-level judgment conditions are classified into the second-level candidate parameter subset. For each parameter in the second-level candidate parameter subset, its coupling influence weight and cross-condition coverage ratio are similarly retrieved and compared with the preset second-level weight threshold and preset second-level coverage threshold in the second-level parameter judgment conditions to determine whether the parameter simultaneously meets the two threshold requirements, generating a second-level judgment result. Based on this result, key acoustic and vibration parameters that meet the second-level judgment conditions are classified into the second-level parameter subset, while parameters that do not meet the second-level judgment conditions are removed. The selected first-level parameter subset and second-level parameter subset are summarized and integrated to form a graded acoustic and vibration parameter set.

[0113] In one embodiment, based on a subset of secondary parameters, a multi-objective collaborative optimization is performed on the preliminary ship compartment design scheme to obtain an optimized ship compartment design scheme, including:

[0114] S61, Determine the optimization index of the scheme based on the key acoustic and vibration parameters in the secondary parameter subset;

[0115] S62, based on the scheme optimization indicators and combined with the preset ship compartment design specifications and manufacturing requirements, sets the optimization objectives and constraints for multi-objective collaborative optimization;

[0116] S63. Based on the optimization objectives and constraints, the preliminary ship cabin design scheme is adjusted through a preset neural network model to form an optimized ship cabin design scheme.

[0117] Specifically, in implementation, key acoustic and vibration parameters in the secondary parameter subset (such as damping coefficients of secondary cabin components, density of local sound insulation materials, and connection stiffness of components in non-core areas) can be used, combined with the current acoustic and vibration performance shortcomings of the preliminary ship cabin design (such as excessive sound pressure levels in local areas of the cabin and excessive vibration acceleration in specific frequency bands), to determine the optimization indicators of the scheme. For example, reducing the sound pressure level in the corner areas of the living quarters to below 55 dB and controlling the vibration acceleration of the equipment bay's auxiliary supports to within 0.1g (g is the acceleration due to gravity) can be used as specific optimization indicators. Based on the determined optimization indicators, and in conjunction with the pre- The design specifications for ship compartments (such as industry standards like "Requirements for Acoustic Design of Ship Compartments" and "Ship Structural Strength Specifications") and manufacturing requirements (such as material procurement feasibility, upper limit of processing precision, and manufacturing cost budget range) are established. Multi-objective collaborative optimization objectives and constraints are set. In addition to acoustic and vibration performance-related objectives, optimization objectives include structural strength objectives (such as a bending strength of bulkhead components not less than 235 MPa) and manufacturing cost objectives (such as a manufacturing cost increase of no more than 5% after optimization of a single compartment). Constraints include constraints on the range of secondary parameter values ​​(such as a sound insulation material laying density of not less than 20 kg / m²). 2 And not higher than 50kg / m 2 To avoid issues such as excessively low density affecting sound insulation or excessively high density increasing structural load, structural dimensional constraints (e.g., secondary component thickness not less than 8mm to meet foundation load requirements), and manufacturing process constraints (e.g., component connection methods must be compatible with existing production line welding processes), the structural parameters and current values ​​of secondary parameters of the preliminary ship cabin design scheme, along with the set optimization objectives and constraints, are used as input data and imported into a preset neural network model. The model iteratively adjusts the values ​​of secondary parameters through a backpropagation algorithm and simultaneously verifies the degree to which the adjusted scheme meets each optimization objective and whether it conforms to the constraints. When the adjusted scheme simultaneously meets all optimization objectives and does not break the constraints, the iteration stops, and the adjusted scheme is determined as the optimized ship cabin design scheme.

[0118] The aforementioned intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments acquires structural data and acoustic-vibration data of the ship compartment to be optimized under multiple operating conditions. An initial simulation model is constructed using the finite element analysis method. Based on this model, data on the propagation paths and attenuation characteristics of acoustic-vibration energy under multiple operating conditions are simulated, and an acoustic-vibration coupling matrix is ​​constructed. Based on this matrix, key acoustic-vibration parameter sets under multiple operating conditions are identified and classified into primary and secondary parameter subsets according to preset screening rules. A preliminary ship compartment design scheme is generated by combining the initial simulation model and the primary parameter subsets through a preset neural network model. Based on the secondary parameter subsets and in conjunction with preset ship compartment design specifications and manufacturing requirements, an optimization objective is set. By performing multi-objective collaborative optimization of the preliminary scheme under constraints, this approach solves the technical problems of traditional design methods that rely on empirical formula derivation or simplified simulation under single working conditions, and fail to fully consider the complex propagation and coupling characteristics of acoustic and vibration energy under multiple working conditions. These problems result in poor acoustic and vibration performance stability, low design accuracy, low optimization efficiency, and difficulty in balancing cabin acoustic and vibration performance, structural strength, and manufacturing costs. This approach enables accurate discovery of the acoustic and vibration coupling law of ship cabins under multiple working conditions, improves the scientific nature and optimization efficiency of ship cabin design, ensures that the optimized design scheme has excellent acoustic and vibration performance under multiple working conditions, and simultaneously takes into account structural strength and manufacturing cost control, thereby improving the overall design quality and user experience of ship cabins.

[0119] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0120] Based on the same inventive concept, this application also provides an intelligent optimization design system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments for implementing the aforementioned intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the intelligent optimization design system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments provided below can be found in the limitations of the intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments described above, and will not be repeated here.

[0121] In one exemplary embodiment, such as Figure 2 As shown, an intelligent optimization design system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments is provided, including:

[0122] The compartment data acquisition module 101 is used to acquire structural data of the ship compartment to be optimized and acoustic and vibration data under multiple operating conditions.

[0123] The simulation model construction module 102 is used to construct an initial simulation model based on structural data and acoustic vibration data using the finite element analysis method.

[0124] The acoustic vibration characteristic simulation module 103 is used to simulate and obtain acoustic vibration energy propagation path and attenuation characteristic data under multiple working conditions based on the initial simulation model.

[0125] The relation matrix construction module 104 is used to construct the acoustic-vibration coupling relation matrix of the initial simulation model based on the acoustic-vibration energy propagation path and attenuation characteristic data.

[0126] The key parameter identification module 105 is used to identify the key acoustic and vibration parameter set under multiple working conditions based on the acoustic-vibration coupling relationship matrix.

[0127] The parameter classification and filtering module 106 is used to classify and filter the key acoustic and vibration parameter set according to the preset filtering rules to form a graded acoustic and vibration parameter set, which includes a primary parameter subset and a secondary parameter subset.

[0128] Design scheme generation module 107 is used to combine the initial simulation model and the first-level parameter subset to generate a preliminary ship compartment design scheme through a preset neural network model.

[0129] The scheme collaborative optimization module 108 is used to perform multi-objective collaborative optimization on the preliminary ship compartment design scheme based on the secondary parameter subset, so as to obtain the optimized ship compartment design scheme.

[0130] In one embodiment, the acoustic vibration characteristic simulation module 103 is further configured to:

[0131] Based on acoustic and vibration data under multiple working conditions, the structural constraints and acoustic and vibration excitation conditions corresponding to each working condition are determined in the initial simulation model.

[0132] Based on structural constraints and acoustic vibration excitation conditions, an initial simulation model is used to perform acoustic vibration response simulation calculations, and the acoustic vibration response simulation results are obtained.

[0133] From the acoustic vibration response simulation results, the acoustic vibration characteristic locations where acoustic vibration energy converges and is transferred are selected;

[0134] Based on the acoustic vibration characteristic locations, the temporal variation data of acoustic vibration energy at each characteristic location in the acoustic vibration response simulation results are extracted;

[0135] The structural connection method and medium energy transfer law of the acoustic vibration characteristic location are obtained, and the acoustic vibration characteristic location, structural connection method and medium energy transfer law are integrated to form energy propagation nodes;

[0136] By combining time-series variation data with energy propagation nodes, the transmission trajectory of acoustic and vibration energy from the excitation source to the ship's compartments is traced in reverse to obtain acoustic and vibration energy propagation path data;

[0137] For each path in the acoustic vibration energy propagation path data, the energy transfer is divided into different stages;

[0138] Based on the acoustic vibration response simulation results, the energy amplitude attenuation and attenuation rate of the acoustic vibration energy propagation path in each stage are calculated to obtain attenuation characteristic data.

[0139] In one embodiment, the relation matrix construction module 104 is further configured to:

[0140] The number of rows and columns of the matrix to be constructed is determined based on the energy propagation nodes in the acoustic energy propagation path data.

[0141] Based on the energy propagation nodes and the number of rows and columns in the matrix, establish a unique correspondence between each energy propagation node and the matrix row and column indices;

[0142] Traverse the acoustic energy propagation path data and, based on the unique correspondence, record the matrix row and column index combination of adjacent energy propagation nodes on each path;

[0143] For each combination of matrix row and column indices, the acoustic-vibration coupling coefficient is calculated based on the attenuation characteristic data;

[0144] Fill the acoustic-vibration coupling coefficients into the cells of the corresponding row and column indices in the matrix to be constructed, and assign preset initial values ​​to the cells of the matrix to be constructed that have not been filled with data, thereby generating the acoustic-vibration coupling relationship matrix.

[0145] In one embodiment, the relation matrix construction module 104 calculates the acoustic-vibration coupling relation matrix using the following formula:

[0146]

[0147] in, C is the acoustic-vibration coupling matrix, and n is the total number of nodes for acoustic-vibration energy propagation. ij E represents the acoustic-vibration energy coupling coefficient from node i to node j. trans，ij E represents the acoustic energy transmitted from node i to node j. soure，i d represents the initial acoustic energy value of node i. ij β represents the effective propagation distance between node i and node j, and β represents the energy attenuation coefficient.

[0148] In one embodiment, the key parameter identification module 105 is further configured to:

[0149] For each working condition in the multi-condition scenario, the coupling influence weight of each acoustic and vibration parameter under the corresponding working condition is calculated based on the acoustic-vibration coupling relationship matrix.

[0150] The coupling influence weights are compared with preset weight thresholds, and acoustic vibration parameters with coupling influence weights greater than preset weight thresholds are selected to form a subset of candidate acoustic vibration parameters for each working condition.

[0151] Within the subset of candidate acoustic and vibration parameters corresponding to each working condition, the cross-working condition coverage ratio of each acoustic and vibration parameter is statistically analyzed.

[0152] The cross-condition coverage ratio is compared with a preset ratio threshold, and acoustic and vibration parameters with a cross-condition coverage ratio greater than the preset ratio threshold are selected to form a set of key acoustic and vibration parameters under multiple conditions.

[0153] In one embodiment, the preset filtering rules in the parameter hierarchical filtering module 106 include primary parameter judgment conditions and secondary parameter judgment conditions; wherein, the primary parameter judgment condition is that the coupling influence weight is greater than or equal to the preset primary weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset primary coverage threshold; the secondary parameter judgment condition is that the coupling influence weight is greater than or equal to the preset secondary weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset secondary coverage threshold; the preset primary weight threshold is greater than the preset secondary weight threshold, and the preset primary coverage threshold is greater than the preset secondary coverage threshold;

[0154] The key acoustic and vibration parameter sets are classified and filtered according to preset screening rules to form a graded acoustic and vibration parameter set, including:

[0155] Determine whether each key acoustic and vibration parameter meets the first-level parameter judgment condition, generate the first-level judgment result, and based on the first-level judgment result, classify the key acoustic and vibration parameters that meet the first-level parameter judgment condition into the first-level parameter subset, and classify the key acoustic and vibration parameters that do not meet the first-level parameter judgment condition into the second-level candidate parameter subset.

[0156] Determine whether the key acoustic and vibration parameters in the secondary candidate parameter subset meet the secondary parameter judgment conditions, generate secondary judgment results, and classify the key acoustic and vibration parameters that meet the secondary parameter judgment conditions into the secondary parameter subset based on the secondary judgment results;

[0157] The primary and secondary parameter subsets are integrated to form a hierarchical acoustic and vibration parameter set.

[0158] In one embodiment, the scheme collaborative optimization module 108 is further configured to:

[0159] Based on the key acoustic and vibration parameters in the secondary parameter subset, the optimization indicators of the scheme are determined;

[0160] Based on the scheme optimization indicators, and combined with the preset ship compartment design specifications and manufacturing requirements, the optimization objectives and constraints of multi-objective collaborative optimization are set.

[0161] Based on the optimization objectives and constraints, the preliminary ship cabin design scheme is adjusted using a pre-set neural network model to form an optimized ship cabin design scheme.

[0162] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship compartments as described above.

[0163] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0164] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0165] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. An intelligent optimization design method based on multi-dimensional acoustic-vibration coupling simulation of ship cabins, characterized in that, The method includes: Acquire structural data and acoustic and vibration data under multiple operating conditions for the ship's compartments to be optimized; Based on the structural data and the acoustic and vibration data, an initial simulation model is constructed using the finite element analysis method. Based on the initial simulation model, data on the propagation path and attenuation characteristics of acoustic and vibration energy under multiple operating conditions are simulated and obtained. Based on the acoustic and vibration energy propagation path and attenuation characteristic data, the acoustic and vibration coupling relationship matrix of the initial simulation model is constructed; Based on the aforementioned acoustic-vibration coupling matrix, the set of key acoustic-vibration parameters under multiple operating conditions is identified. The key acoustic and vibration parameter set is classified and filtered according to preset filtering rules to form a graded acoustic and vibration parameter set, which includes a primary parameter subset and a secondary parameter subset. By combining the initial simulation model with the first-level parameter subset, a preliminary ship cabin design scheme is generated through a preset neural network model; Based on the aforementioned subset of secondary parameters, the preliminary ship cabin design scheme is subjected to multi-objective collaborative optimization to obtain an optimized ship cabin design scheme.

2. The method according to claim 1, characterized in that, The step of simulating and obtaining acoustic and vibration energy propagation paths and attenuation characteristics under multiple operating conditions based on the initial simulation model includes: Based on the acoustic and vibration data under the multiple working conditions, the structural constraints and acoustic and vibration excitation conditions corresponding to each working condition are determined in the initial simulation model. Based on the structural constraints and the acoustic vibration excitation conditions, the acoustic vibration response simulation calculation is performed using the initial simulation model to obtain the acoustic vibration response simulation results. From the acoustic vibration response simulation results, the acoustic vibration characteristic locations where acoustic vibration energy converges and is transferred are selected; Based on the acoustic vibration feature locations, extract the time-series variation data of acoustic vibration energy at each feature location in the acoustic vibration response simulation results; The structural connection method and medium energy transfer law of the acoustic vibration characteristic location are obtained, and the acoustic vibration characteristic location, the structural connection method and the medium energy transfer law are integrated to form an energy propagation node; By combining the time-series change data with the energy propagation nodes, the transmission trajectory of acoustic energy from the excitation source to the ship's cabin is traced in reverse to obtain the acoustic energy propagation path data; For each path in the acoustic energy propagation path data, the energy transfer is divided into different stages; Based on the acoustic vibration response simulation results, the energy amplitude attenuation and attenuation rate of the acoustic vibration energy propagation path in each stage are calculated to obtain the attenuation characteristic data.

3. The method according to claim 2, characterized in that, The acoustic-vibration coupling matrix of the initial simulation model, constructed based on the acoustic-vibration energy propagation path and attenuation characteristic data, includes: The number of rows and columns of the matrix to be constructed is determined based on the energy propagation nodes in the acoustic energy propagation path data. Based on the energy propagation nodes and the number of rows and columns of the matrix, establish a unique correspondence between each energy propagation node and the matrix row and column indexes; Traverse the acoustic energy propagation path data and, in conjunction with the unique correspondence, record the matrix row and column index combination of adjacent energy propagation nodes on each path; For each combination of matrix row and column indices, the acoustic-vibration coupling coefficient is calculated based on the attenuation characteristic data; The acoustic-vibration coupling coefficient is filled into the cell corresponding to the row and column index in the matrix to be constructed, and a preset initial value is assigned to the cells in the matrix to be constructed that have not been filled with data, thereby generating the acoustic-vibration coupling relationship matrix.

4. The method according to claim 3, characterized in that, The acoustic-vibration coupling matrix is ​​calculated using the following formula: in, C is the acoustic-vibration coupling matrix, and n is the total number of nodes for acoustic-vibration energy propagation. ij E represents the acoustic-vibration energy coupling coefficient from node i to node j. trans，ij E represents the acoustic energy transmitted from node i to node j. soure，i d represents the initial acoustic energy value of node i. ij β represents the effective propagation distance between node i and node j, and β represents the energy attenuation coefficient.

5. The method according to claim 1, characterized in that, The identification of key acoustic and vibration parameter sets under multiple operating conditions based on the acoustic-vibration coupling relationship matrix includes: For each of the multiple operating conditions, the coupling influence weight of each acoustic and vibration parameter under the corresponding operating condition is calculated based on the acoustic-vibration coupling relationship matrix. The coupling influence weights are compared with preset weight thresholds to filter out acoustic vibration parameters whose coupling influence weights are greater than the preset weight thresholds, thus forming a subset of candidate acoustic vibration parameters corresponding to each working condition. Within the subset of candidate acoustic and vibration parameters corresponding to each of the aforementioned working conditions, the cross-working condition coverage ratio of each acoustic and vibration parameter is statistically analyzed. The cross-condition coverage ratio is compared with a preset ratio threshold, and acoustic and vibration parameters with a cross-condition coverage ratio greater than the preset ratio threshold are selected to form the key acoustic and vibration parameter set under the multi-condition.

6. The method according to claim 5, characterized in that, The preset filtering rules include primary parameter judgment conditions and secondary parameter judgment conditions; wherein, the primary parameter judgment condition is that the coupling influence weight is greater than or equal to the preset primary weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset primary coverage threshold; the secondary parameter judgment condition is that the coupling influence weight is greater than or equal to the preset secondary weight threshold, and the cross-working condition coverage ratio is greater than or equal to the preset secondary coverage threshold; the preset primary weight threshold is greater than the preset secondary weight threshold, and the preset primary coverage threshold is greater than the preset secondary coverage threshold; The step of classifying and filtering the key acoustic and vibration parameter set according to preset filtering rules to form a graded acoustic and vibration parameter set includes: Determine whether each of the key acoustic and vibration parameters meets the first-level parameter judgment conditions, generate a first-level judgment result, and based on the first-level judgment result, classify the key acoustic and vibration parameters that meet the first-level parameter judgment conditions into the first-level parameter subset, and classify the key acoustic and vibration parameters that do not meet the first-level parameter judgment conditions into the second-level candidate parameter subset. Determine whether the key acoustic and vibration parameters in the subset of secondary candidate parameters meet the secondary parameter judgment conditions, generate secondary judgment results, and classify the key acoustic and vibration parameters that meet the secondary parameter judgment conditions into the secondary parameter subset based on the secondary judgment results; The first-level parameter subset and the second-level parameter subset are integrated to form the graded acoustic and vibration parameter set.

7. The method according to claim 1, characterized in that, The step of performing multi-objective collaborative optimization on the preliminary ship compartment design scheme based on the secondary parameter subset to obtain an optimized ship compartment design scheme includes: Based on the key acoustic and vibration parameters in the secondary parameter subset, the scheme optimization index is determined; Based on the optimization indicators of the above scheme, and in combination with the preset ship compartment design specifications and manufacturing requirements, the optimization objectives and constraints of the multi-objective collaborative optimization are set. Based on the optimization objective and the constraints, the preliminary ship cabin design scheme is adjusted using the preset neural network model to form the optimized ship cabin design scheme.

8. An intelligent optimization design system based on multi-dimensional acoustic-vibration coupling simulation of ship compartments, characterized in that, The system includes: The compartment data acquisition module is used to acquire structural data and acoustic and vibration data of the compartments of the ship to be optimized under multiple operating conditions; The simulation model construction module is used to construct an initial simulation model based on the structural data and the acoustic vibration data using the finite element analysis method. The acoustic vibration characteristic simulation module is used to simulate and obtain acoustic vibration energy propagation path and attenuation characteristic data under multiple working conditions based on the initial simulation model. The relation matrix construction module is used to construct the acoustic-vibration coupling relation matrix of the initial simulation model based on the acoustic-vibration energy propagation path and attenuation characteristic data. The key parameter identification module is used to identify the set of key acoustic and vibration parameters under multiple working conditions based on the acoustic-vibration coupling relationship matrix. The parameter hierarchical filtering module is used to perform hierarchical filtering of the key acoustic and vibration parameter set according to preset filtering rules to form a hierarchical acoustic and vibration parameter set, which includes a primary parameter subset and a secondary parameter subset. The design scheme generation module is used to combine the initial simulation model with the first-level parameter subset and generate a preliminary ship cabin design scheme through a preset neural network model. The scheme collaborative optimization module is used to perform multi-objective collaborative optimization on the preliminary ship cabin design scheme based on the secondary parameter subset, so as to obtain an optimized ship cabin design scheme.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images