A compound environment physical simulation method and system for river-sea direct ship type optimization

Abstract

The application discloses a composite environment physical simulation method and system for optimization of a river-sea direct ship type, and relates to the technical field of ship design and testing. The method comprises the following steps: obtaining a test database of a river-sea direct ship; the test database comprises test results obtained by using a composite physical environment simulation method to perform ship pool performance testing based on a ship model; the ship pool performance testing comprises a ship model shallow water test, a ship model deep water test and a ship model shallow-deep transition zone test; performing simulation parameterization calculation on the river-sea direct ship to establish a simulation database; constructing a virtual-agent model of the river-sea direct ship based on the test database and the simulation database; and based on the virtual-agent model, using an optimization algorithm to optimize ship type design parameters of the river-sea direct ship with the optimal multi-performance index as the target, so as to obtain optimal ship type design parameters of the river-sea direct ship. The application can improve the reliability and optimization efficiency of the ship type design of the river-sea direct ship.

Description

Technical Field

[0001] This application relates to the field of ship design and testing technology, and in particular to a composite environmental physical simulation method and system for optimizing the hull form of river-sea direct vessels. Background Technology

[0002] River-sea direct vessels are a type of transport that can navigate directly between inland waterways and near-shore waters without transshipment at ports, offering advantages such as high transport efficiency, low cost, and flexible operation. Compared to traditional seagoing or river vessels, the most significant characteristic of river-sea direct vessels is that they must simultaneously meet the requirements of shallow draft, good maneuverability, and low resistance for inland waterway navigation, as well as the requirements of stability, safety, and seakeeping for near-shore navigation.

[0003] River-sea direct vessels frequently traverse river sections and sea sections, facing significantly different environmental conditions: river sections often have limited water depth, narrow channels, and complex currents, resulting in shallow-water effects and limited maneuverability; while sea sections present deep water, large waves, and irregular wave environments, requiring the hull to withstand significant wave loads and structural impacts. These navigation characteristics place stringent requirements on the hull design of river-sea direct vessels.

[0004] For a long time, pool testing has been the most effective and reliable method for verifying the performance of river-sea direct-route vessels. Currently, pool tests of river-sea direct-route vessels in river and sea sections typically need to be conducted separately in different test pools or simulated environments. On the one hand, it is impossible to continuously simulate river and sea sections in the same test, lacking the ability to reproduce transitional environments; on the other hand, it is impossible to achieve a comprehensive performance evaluation of the designed vessel in a combined river-sea environment, which significantly limits the efficiency of verification and optimization of river-sea direct-route vessels.

[0005] Therefore, improving the reliability and optimization efficiency of the ship design for direct river-sea vessels has become an urgent problem to be solved. Summary of the Invention

[0006] The purpose of this application is to provide a composite environmental physical simulation method and system for optimizing the hull form of river-sea direct vessels, which can improve the reliability and optimization efficiency of the hull form design of river-sea direct vessels.

[0007] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels, including: Obtain a test database for river-sea direct vessels; the test database includes: test results obtained from ship tank performance tests based on ship models using a composite physical environment simulation method; the ship tank performance tests include: shallow water tests of ship models, deep water tests of ship models, and shallow-deep transition zone tests of ship models. The simulated parameterization calculations were performed on the aforementioned river-sea direct vessel to establish a simulation database; the experimental database includes: simulation results obtained from simulation performance tests based on the ship model using simulation methods; Based on the experimental database and the simulation database, a virtual-proxy model of the river-sea direct vessel is constructed; the virtual-proxy model is used to predict various performance indicators of river-sea direct vessels with different ship type design parameters. Based on the aforementioned virtual-agent model, with the goal of optimizing multiple performance indicators, an optimization algorithm is used to optimize the ship design parameters of the river-sea direct vessel, thereby obtaining the optimal ship design parameters of the river-sea direct vessel.

[0008] In one embodiment, before obtaining the experimental database of river-sea direct vessels, the composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels further includes: Obtain the initial ship type for direct river-sea vessels; Based on the initial ship design, a ship model is obtained through design and fabrication. The test environment of the ship model is determined by a composite physical environment simulation method. The test environment includes: shallow water environment, deep water environment and shallow-to-deep transition environment simulated by the test device. The test device includes: a test tank and a pluggable shoal module, a bottom lifting platform and a current generation system installed in the test tank. The resistance performance and maneuverability of the ship model were tested in shallow water environment to obtain the shallow water test results. Based on the deep-water environment, a deep-water test was conducted on the stability and seakeeping of the ship model to obtain the deep-water test results. Based on the shallow-deep transition environment, a shallow-deep transition zone test was conducted on the ship model to assess its stability and maneuverability risks, and the results of the transition zone test were obtained. A test database is constructed based on the results of the shallow water test, the deep water test, and the transition zone test.

[0009] In one embodiment, a ship model is obtained by designing and manufacturing based on an initial ship form, specifically including: The geometric scaling ratio is determined based on the length of the test tank, the speed of the towing vehicle, and the wave-generating capability. The actual ship form is determined based on the initial ship form. The actual ship lines are converted into ship model lines according to the geometric scaling ratio; The ship model is obtained by processing according to the aforementioned ship model line drawing.

[0010] In one embodiment, a test database is constructed based on the shallow water test results, the deep water test results, and the transition zone test results, specifically including: The results of the shallow water test, the deep water test, and the transition zone test are combined to obtain the comprehensive test results. Based on engineering experience, the results of the comprehensive test that do not meet the performance requirements are modified, and the modified comprehensive test results are determined as the test database.

[0011] In one embodiment, a virtual-agent model of the river-sea direct vessel is constructed based on the experimental database and the simulation database, specifically including: The experimental database and the simulation database are merged to obtain a fused database; The fused database is preprocessed to obtain the training set; The ship design parameters of the ship model are used as input, and the corresponding performance test results in the training set are used as output to train the machine learning model, thereby obtaining the virtual-agent model of the river-sea direct vessel; each virtual-agent model corresponds to the prediction of a performance index.

[0012] In one embodiment, based on the virtual-agent model, with the goal of optimizing multiple performance indicators, an optimization algorithm is used to optimize the hull design parameters of the river-sea direct vessel, obtaining the optimal hull design parameters of the river-sea direct vessel, specifically including: Construct the fitness function of the genetic optimization algorithm based on the virtual-proxy model; With the fitness function as the objective, the hull design parameters of the river-sea direct vessel are iteratively optimized multiple times to obtain the optimal hull design parameters of the river-sea direct vessel.

[0013] In one embodiment, the fitness function is expressed as: ; in, Indicates ship design parameters The corresponding fitness function value; This represents the first performance metric value output by the first virtual-agent model. This represents the second performance metric value output by the second virtual-agent model; Indicates the first The output of the virtual-agent model is the first Various performance index values; This represents the weighting coefficient corresponding to the first performance index value; This represents the weighting coefficient corresponding to the second performance index value; Indicates the first The weighting coefficients corresponding to the performance index values.

[0014] In one embodiment, the machine learning model is a support vector machine.

[0015] In one embodiment, the ship design parameters include: length, beam, draft, and bow and stern shapes.

[0016] Secondly, this application provides a composite environmental physical simulation system for optimizing the hull form of river-sea direct vessels, comprising: The test database acquisition module is used to acquire the test database of the river-sea direct vessel; the test database includes: test results obtained from the performance testing of the ship in the water tank based on the ship model and using a composite physical environment simulation method; the performance testing of the ship in the water tank includes: shallow water test of the ship model, deep water test of the ship model, and shallow-deep transition zone test of the ship model. The simulation database construction module is used to perform simulation parameterization calculations on the river-sea direct vessel and establish a simulation database; the experimental database includes: simulation results obtained by simulation performance testing based on the ship model using simulation methods. The virtual-proxy model construction module is used to construct a virtual-proxy model of the river-sea direct vessel based on the experimental database and the simulation database; the virtual-proxy model is used to predict various performance indicators of river-sea direct vessels with different ship type design parameters; The multi-objective optimization module is used to optimize the ship design parameters of the river-sea direct vessel based on the virtual-agent model, with the goal of optimizing multiple performance indicators, and to obtain the optimal ship design parameters of the river-sea direct vessel.

[0017] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a composite environmental physical simulation method and system for optimizing the hull form of river-sea direct vessels. Based on a ship model, a composite physical environment simulation method is used to conduct performance tests in the ship's water tank, thereby constructing a test database for river-sea direct vessels. The ship's water tank performance tests include: shallow water tests, deep water tests, and shallow-to-deep transition zone tests of the ship model. The test database constructed using the composite physical environment simulation method can realistically reproduce the combined river-sea navigation environment, avoiding the problems of separate tests in different test tanks and inconsistent data in the past. It can also capture the special effects of the shallow-to-deep transition zone, which can improve the reliability of hull design when applied to hull design. Combined with the simulation database, a virtual-surrogate model of the river-sea direct vessel is constructed. Based on the virtual-surrogate model and optimization algorithm, with the goal of optimizing multiple performance indicators, the hull design parameters are optimized. This can quickly obtain hull design parameters that meet the dual navigation requirements of river and sea, improving the optimization efficiency of the hull design of river-sea direct vessels, thereby helping to improve the navigation performance of river-sea direct vessels and enhance their overall economic benefits. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the 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.

[0019] Figure 1 A flowchart illustrating a composite environmental physical simulation method for optimizing the hull form of a river-sea direct vessel, provided in an embodiment of this application; Figure 2 A detailed implementation process diagram of a composite environmental physical simulation method for optimizing the hull form of a river-sea direct vessel, provided in another embodiment of this application; Figure 3 This is a schematic diagram of the functional modules of a composite environmental physical simulation system for optimizing the hull shape of a river-sea direct vessel, provided as another embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] In one exemplary embodiment, such as Figure 1 As shown, a composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels is provided, including: Step 101: Obtain the test database for the river-sea direct vessel; the test database includes: test results obtained from the performance testing of the vessel in the water tank based on the ship model and using a composite physical environment simulation method; the performance testing of the vessel in the water tank includes: shallow water test of the ship model, deep water test of the ship model, and shallow-deep transition zone test of the ship model.

[0023] Step 102: Perform simulation parameterization calculations on the river-sea direct vessel and establish a simulation database; the experimental database includes: simulation results obtained by simulation performance testing based on the ship model using simulation methods.

[0024] Step 103: Based on the experimental database and the simulation database, construct a virtual-proxy model for the river-sea direct vessel; the virtual-proxy model is used to predict various performance indicators of river-sea direct vessels with different ship type design parameters.

[0025] Step 104: Based on the virtual-agent model, with the goal of optimizing multiple performance indicators, the ship design parameters of the river-sea direct vessel are optimized using an optimization algorithm to obtain the optimal ship design parameters of the river-sea direct vessel.

[0026] In one embodiment, the ship design parameters include: ship length, beam, draft, bow and stern shape, and other characterization parameters.

[0027] In one embodiment, the composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels further includes: constructing an experimental database for river-sea direct vessels, specifically: (1) Obtain the initial ship type of the river-sea direct vessel. Specifically: Obtain the initial ship type of the river-sea direct vessel through the master model or design principles.

[0028] (2) Based on the initial ship type, design and process to obtain the ship model, and prepare for the ship model tank test. The preparation work for the ship model tank test includes, but is not limited to: collecting prototype ship type data, selecting the scale ratio, designing the scaled model, drawing ship model lines, and designing test functions.

[0029] This step specifically includes: determining the geometric scale ratio based on the length of the test tank, the speed of the towing vehicle, and the wave-generating capability; determining the actual ship shape based on the initial ship shape; converting the actual ship shape into a ship model shape drawing according to the geometric scale ratio; and processing the ship model according to the ship model shape drawing to obtain the ship model.

[0030] (3) The test environment of the ship model is determined by a composite physical environment simulation method. The test environment includes: shallow water environment, deep water environment and shallow-deep transition environment simulated by the test device. The test device includes: test tank and pluggable shoal module, bottom lifting platform and flow generation system installed in the test tank.

[0031] In this step, a composite physical environment simulation method is used to conduct performance tests on ships in the water tank. The main function of the composite physical environment simulation method is to reproduce the diverse environments faced by river-sea direct ships. Specifically, it refers to the special navigation scenarios in which river-sea direct ships switch between river sections (shallow water, narrow channels, complex flow fields) and sea sections (deep water, large waves, long period waves), thereby revealing the laws of ship performance change and establishing unified performance evaluation conditions.

[0032] The composite physical environment simulation method uses components such as pluggable shoal modules, bottom lifting platforms, and flow generation systems to simulate the composite shallow water environment, deep water environment, and shallow-to-deep transition environment of river-sea direct vessels. The composite physical environment of river-sea direct vessels includes shallow water environment, deep water environment, and shallow-to-deep transition environment.

[0033] (4) Based on the shallow water environment, a shallow water test is conducted on the resistance performance and maneuverability of the ship model to obtain the shallow water test results; based on the deep water environment, a deep water test is conducted on the stability and seakeeping performance of the ship model to obtain the deep water test results; based on the shallow-deep transition environment, a shallow-deep transition zone test is conducted on the stability risk and maneuverability risk of the ship model to obtain the transition zone test results.

[0034] In practical applications, this step involves conducting shallow-water tests, deep-water tests, and shallow-to-deep transition zone tests on the ship model, based on the experimental environment provided by the composite physical environment simulation method. The main contents of the shallow-water tests include: resistance and propulsion characteristic testing, maneuverability testing, and stability testing; the main contents of the deep-water tests include: resistance and speed characteristic testing, seakeeping testing, and stability testing; the shallow-to-deep transition zone tests are responsible for: determining the resistance and maneuverability characteristics of the transition zone, analyzing the transient response of the hull motion (such as sudden changes in trim / heel), and observing the yaw and maneuverability of the ship model.

[0035] (5) Construct a test database based on the shallow water test results, the deep water test results, and the transition zone test results. Specifically, this includes: The shallow water test results, deep water test results, and transition zone test results are integrated. Based on the test results, the overall performance of the designed vessel in the river section and sea section is evaluated to obtain the comprehensive test results. Based on engineering experience, the results in the comprehensive test results that do not meet the performance requirements are modified, and the modified comprehensive test results are determined as the test database.

[0036] In one embodiment, step 103 combines the experimental database and the simulation database to establish a virtual-proxy model for river-sea direct transport vessels. The main purpose of the virtual-proxy model is to integrate experimental and simulation data, leverage their complementary advantages, form a performance prediction model that is closer to the actual ship operating conditions, and improve the mapping efficiency between ship type and comprehensive performance parameters. It will then be used in the ship type optimization process.

[0037] In the process of ship optimization, it is usually necessary to iterate repeatedly between multiple objectives (resistance, seakeeping, stability, maneuverability, etc.). Directly using CFD or actual tests will result in excessive computation / experimentation. After establishing a surrogate model, performance can be predicted quickly within the design space, thus achieving rapid optimization.

[0038] The process of establishing a virtual-agent model includes: database construction and management, data fusion and preprocessing, agent model establishment, model training and validation, and application optimization.

[0039] Step 103 specifically includes: The experimental database and the simulation database are merged to obtain a fused database.

[0040] The fused database is preprocessed to obtain the training set.

[0041] Using the ship model's hull design parameters as input and the corresponding performance test results from the training set as output, the machine learning model is trained to obtain a virtual-proxy model of the river-sea direct vessel; each virtual-proxy model corresponds to the prediction of one performance index. The machine learning model can be a support vector machine.

[0042] In one embodiment, step 104 specifically includes: The fitness function of the genetic optimization algorithm is constructed based on the virtual-proxy model. With the goal of maximizing the fitness function (i.e. approximating the Pareto front curve), the hull design parameters of the river-sea direct vessel are iteratively optimized multiple times to obtain the optimal hull design parameters of the river-sea direct vessel.

[0043] The fitness function is expressed as follows: ; in, Indicates ship design parameters The corresponding fitness function value; This represents the first performance metric value output by the first virtual-agent model. This represents the second performance metric value output by the second virtual-agent model; Indicates the first The output of the virtual-agent model is the first Various performance index values; This represents the weighting coefficient corresponding to the first performance index value; This represents the weighting coefficient corresponding to the second performance index value; Indicates the first The weighting coefficients corresponding to the performance index values.

[0044] This step, based on the established virtual-proxy model, further establishes a multi-objective optimization design process for river-sea direct vessels centered around a genetic optimization algorithm. After constructing the experimental database, designers manually modify vessel design schemes that do not meet performance requirements based on engineering experience; this is one optimization process. Here, the genetic algorithm optimization process is used to obtain vessel design schemes that approximate the Pareto front curve; this is the second optimization process.

[0045] Modifying the ship design means restarting the design and fabrication of the ship model, as well as conducting new tank tests on the ship model, which will result in higher design costs. The addition of the genetic algorithm process is beneficial to improving the efficiency of the optimization design of river-sea direct vessels and reducing design costs.

[0046] The virtual-agent model is not limited to the design of one type or class of ships. With the accumulation of design cases and experimental and simulation databases, the ship optimization process proposed in this embodiment will be further improved in terms of reliability and efficiency.

[0047] The above embodiments describe a composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels. This method obtains the initial hull form of the vessel through a parent model or design principles; it then conducts model design and fabrication, preparing for tank testing; performs tank performance testing using a composite physical environment simulation method; sequentially conducts shallow-water, deep-water, and shallow-to-deep transition zone tests on the model within the test environment provided by the simulation method; evaluates the comprehensive performance of the designed hull form across river and sea sections based on the test results; establishes a virtual-surrogate model of the river-sea direct vessel hull form by combining the test and simulation databases; and further develops a multi-objective optimization design process for the river-sea direct vessel hull form based on the established virtual-surrogate model and a genetic optimization algorithm. This method can realistically reproduce the combined river-sea navigation environment, avoiding the previous problems of separate tests in different test tanks and inconsistent data, and can capture the special effects of the shallow-to-deep transition zone. The test environment is closer to the actual ship operating conditions, making the hull design more targeted and practical. Combined with the genetic optimization algorithm and multi-objective optimization method, it can quickly obtain hull form schemes that meet the dual requirements of river and sea navigation.

[0048] The following is a more specific embodiment, combined with Figure 2 The composite environmental physical simulation method described above for optimizing the hull form of direct river-sea vessels will be further introduced in detail.

[0049] See Figure 2 The overall execution process is as follows: (1) Obtain the initial ship type of the river-sea direct vessel through the parent model or design principle.

[0050] (2) Carry out ship model design and processing, and prepare for ship model water tank test.

[0051] Specifically: Based on the length of the pool, the speed of the towing vehicle, and the wave-making capacity, determine the appropriate geometric scaling ratio, convert the actual ship lines into a ship model line drawing according to the scaling ratio, and check the smoothness of the curved surfaces; a wooden ship model can be selected, with the interior hollowed out, ribs or frames added, and the surface sealed and reinforced to ensure waterproofness and strength. At the same time, measure the external dimensions and check the weight, center of gravity, and ballast of the ship model.

[0052] (3) For ship models, the performance of the ship pool is tested based on the composite physical environment simulation method.

[0053] The composite physical environment simulation method is implemented using components such as a pluggable shoal module, a bottom lifting platform, and a flow generation system. The pluggable shoal module is installed in the test tank for rapid setup or removal of the shallow water environment. The bottom lifting platform allows for continuous adjustment of the overall water depth through lifting control. The flow generation system creates a constant or shear flow field within the tank, simulating the inland river flow environment. A wave generator simulates regular and irregular waves in deep water, enabling continuous switching between river sections, sea sections, and river-sea transition zones. The bottom lifting platform, in conjunction with the flow generation system, can flexibly provide shallow or deep water environments for the ship model, meeting its shallow and deep water testing requirements. Furthermore, the bottom lifting platform, combined with the pluggable shoal module, can further achieve shallow and deep water transition effects in different zones of the tank. The above can be adjusted according to actual needs and site conditions during implementation.

[0054] Based on the test environment provided by the composite physical environment simulation method, shallow water test, deep water test and shallow-deep transition zone test of the ship model were carried out in sequence.

[0055] Conduct shallow water tests, using trailer or self-propelled testing, in conjunction with inertial navigation / dynamic capture systems, to determine the resistance of the ship model at different speeds and analyze the shallow water effect; measure turning speed, etc. Z The study investigated the effects of shallow water conditions on rudder performance and course maintenance, including maneuverability.

[0056] Deep-water tests were conducted, with the bottom lifting platform lowered to its maximum depth to ensure the ambient water depth was 4-5 times greater than the model's draft. A wave-making system was used to simulate different sea conditions (regular waves and irregular waves). Sensors (six-component force meter and wave height meter) were set up to record the forces and motion of the model. Data on the ship's heave, pitch, roll, and speed loss under regular and irregular waves were obtained. Waves on the model's deck and bow impact phenomena were analyzed. Stability tests were conducted.

[0057] A stepped underwater environment is created by arranging partitioned lifting plates in the pool to transition from shallow to deep water. The model boat passes through the boundary zone at a constant speed, and resistance, heave, pitch, and roll are recorded in real time. Flow field simulation is introduced to make the hydrodynamics of the transition zone closer to that of a river mouth or estuary.

[0058] (4) Based on the test results, evaluate the comprehensive performance of the designed ship type in the river section and sea section, and construct a test database.

[0059] Ship model tests were conducted sequentially on the river section, the sea section, and the shallow-to-deep transition zone. Based on the numerical simulation results, a unified performance evaluation system was constructed.

[0060] Based on typical operating conditions, key indicators such as resistance, propulsion efficiency, maneuverability, seakeeping, stability, and structural load are determined; each performance indicator is normalized and weighted; and the overall seaworthiness and economy of river-sea direct vessels are realistically reflected under various environmental navigation conditions.

[0061] Determine whether the results of the comprehensive test in the ship model tank meet the performance requirements. If not, make manual modifications until the performance requirements are met, and then construct a test database.

[0062] (5) Conduct simulation parameterization calculations for the designed ship type and establish a simulation database.

[0063] (6) Integrate experimental data with simulation data to establish a virtual-agent model.

[0064] First, multi-source data cleaning and quality assessment are performed, including sensor drift / zero-point calibration and outlier removal; then, confidence weights are assigned to each sample. w Experimental data has a higher weight than simulation data; data with higher repeatability and environmental reproducibility have a higher weight.

[0065] A virtual-surrogate model is built using support vector machines, and the regression function is expressed as: , The design parameter vector for the ship type (length, beam, draft, bow and stern shape, and other characterization parameters, etc.); It is a nonlinear mapping function. w For the weight vector in the feature space, b For bias, For performance metrics.

[0066] The optimization objective of the virtual-agent model is expressed as: .

[0067] C This is the penalty coefficient, used to control model complexity and fault tolerance. and Let be slack variables, where As a positive error slack variable, it is mainly used to handle data points that should have been negative but were misclassified as positive. It is a negative error slack variable used to handle data points that should be positive but are misclassified as negative; N This indicates the number of training samples.

[0068] The constraints are expressed as follows: .

[0069] For input samples; These are the model's predicted values; This represents the tolerance of the regression model, used to indicate the allowable range of error.

[0070] The Gaussian radial kernel function is expressed as: .

[0071] For relative Another set of samples; Adjusting parameters for the kernel function, when When the value is large, the kernel function becomes more sensitive to the distance between samples, while when... When the value is small, the kernel function value changes more gradually.

[0072] The final regression function of the surrogate model is expressed as: .

[0073] in, and It is a dual Lagrange multiplier.

[0074] (7) Based on the established virtual-agent model, a multi-objective optimization design process for river-sea direct vessels based on the genetic optimization algorithm is further established.

[0075] Genetic algorithms, using a global optimization method based on natural selection and genetic mechanisms, address the ship morphology optimization problem for direct river-sea vessels. The fitness function of the genetic optimization algorithm is expressed as: .

[0076] in, For the first i The output of the virtual-agent model is the first i The first performance index value, namely the first i Objective function values, such as drag, stability, and seakeeping performance, etc. w i For the first i The weighting coefficients corresponding to the performance index values ​​reflect the priority of the target.

[0077] The selection probability is expressed as: .

[0078] in, P i For individuals The probability of being selected, where N is the population size.

[0079] The crossover and mutation processes are represented as follows: .

[0080] .

[0081] in, and These represent the new offspring solutions generated by crossover and mutation operations, respectively. and Two parent solutions are needed for crossover or mutation operations; Indicates an interleaving operation; This indicates a mutation operation.

[0082] For ship design schemes that do not meet performance requirements, a ship design scheme that approximates the Pareto front curve is first obtained based on the genetic algorithm optimization process.

[0083] After obtaining the ship design scheme, the ship model design and fabrication process, as well as the ship model tank testing process, are repeated. The design process ends when the optimized design is completed.

[0084] Furthermore, after obtaining the ship design scheme, the designers' engineering experience can be combined to modify and select an optimized design scheme for the river-sea direct vessel that meets the actual engineering and performance requirements. The ship model design and fabrication, as well as the ship model tank testing process, are then re-executed for the revised scheme. This process continues until the optimized design is completed, at which point the design process is finished.

[0085] The above embodiments mainly include a composite physical environment simulation method and a virtual-surrogate model. The composite physical environment simulation method is implemented using components such as a pluggable shoal module, a bottom lifting platform, and a current-generating system. It can sequentially simulate the shallow water, deep water, and shallow-to-deep transition environments that river-sea direct vessels may encounter. Subsequently, based on this, shallow water tests, deep water tests, and shallow-to-deep transition zone tests are conducted on the ship model, accumulating comprehensive performance data such as resistance, maneuverability, seakeeping, and stability of the designed ship type under composite environments. Combined with human experience and experimental and simulation data, a virtual-surrogate model for ship model performance evaluation is established. A multi-objective optimization process for the ship type based on a genetic algorithm is established to complete the optimization of the designed ship type. This embodiment can improve the reliability and optimization efficiency of river-sea direct vessel design, help improve the navigation performance of river-sea direct vessels, and enhance their overall economic benefits.

[0086] This embodiment provides a physical experiment and simulation method that can simulate the river and sea environments on the same physical test platform, supporting rapid iteration and optimization of river-sea direct-route vessels. It can provide technical support for the inland river-sea direct-route vessel industry and further improve its economic benefits.

[0087] Based on the same inventive concept, this application also provides a composite environmental physical simulation system for optimizing the hull shape of river-sea direct vessels, which implements the aforementioned composite environmental physical simulation method for optimizing the hull shape of river-sea direct vessels. 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 composite environmental physical simulation system for optimizing the hull shape of river-sea direct vessels provided below can be found in the limitations of the composite environmental physical simulation method for optimizing the hull shape of river-sea direct vessels described above, and will not be repeated here.

[0088] In one exemplary embodiment, such as Figure 3 As shown, a composite environmental physical simulation system for optimizing the hull form of river-sea direct vessels is provided, comprising: The test database acquisition module 301 is used to acquire the test database of the river-sea direct vessel; the test database includes: test results obtained by using a composite physical environment simulation method to conduct performance tests of the ship in the water tank based on the ship model; the performance tests of the ship in the water tank include: shallow water test of the ship model, deep water test of the ship model, and shallow-deep transition zone test of the ship model.

[0089] The simulation database construction module 302 is used to perform simulation parameterization calculations on the river-sea direct vessel and establish a simulation database; the experimental database includes simulation results obtained by simulation performance testing based on the ship model using simulation methods.

[0090] The virtual-proxy model construction module 303 is used to construct a virtual-proxy model of the river-sea direct vessel based on the experimental database and the simulation database; the virtual-proxy model is used to predict various performance indicators of the river-sea direct vessel with different ship type design parameters.

[0091] The multi-objective optimization module 304 is used to optimize the ship design parameters of the river-sea direct vessel based on the virtual-agent model, with the goal of optimizing multiple performance indicators, and to obtain the optimal ship design parameters of the river-sea direct vessel.

[0092] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0093] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels, characterized in that, The composite environmental physical simulation method for optimizing the hull form of direct river-sea vessels includes: Obtain a test database for river-sea direct vessels; the test database includes: test results obtained from ship tank performance tests based on ship models using a composite physical environment simulation method; the ship tank performance tests include: shallow water tests of ship models, deep water tests of ship models, and shallow-deep transition zone tests of ship models. The simulated parameterization calculations were performed on the aforementioned river-sea direct vessel to establish a simulation database; the experimental database includes: simulation results obtained from simulation performance tests based on the ship model using simulation methods; Based on the experimental database and the simulation database, a virtual-proxy model of the river-sea direct vessel is constructed; the virtual-proxy model is used to predict various performance indicators of river-sea direct vessels with different ship type design parameters. Based on the aforementioned virtual-agent model, with the goal of optimizing multiple performance indicators, an optimization algorithm is used to optimize the ship design parameters of the river-sea direct vessel, thereby obtaining the optimal ship design parameters of the river-sea direct vessel. Before obtaining the experimental database for river-sea direct vessels, the composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels also includes: Obtain the initial ship type for direct river-sea vessels; Based on the initial ship design, a ship model is obtained through design and fabrication. The test environment of the ship model is determined by a composite physical environment simulation method. The test environment includes: shallow water environment, deep water environment and shallow-to-deep transition environment simulated by the test device. The test device includes: a test tank and a pluggable shoal module, a bottom lifting platform and a current generation system installed in the test tank. The resistance performance and maneuverability of the ship model were tested in shallow water environment to obtain the shallow water test results. Based on the deep-water environment, a deep-water test was conducted on the stability and seakeeping of the ship model to obtain the deep-water test results. Based on the shallow-deep transition environment, a shallow-deep transition zone test was conducted on the ship model to assess its stability and maneuverability risks, and the results of the transition zone test were obtained. A test database is constructed based on the results of the shallow water test, the deep water test, and the transition zone test.

2. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 1, characterized in that, Based on the initial ship design, a ship model is obtained through design and fabrication, specifically including: The geometric scaling ratio is determined based on the length of the test tank, the speed of the towing vehicle, and the wave-generating capability. The actual ship form is determined based on the initial ship form. The actual ship lines are converted into ship model lines according to the geometric scaling ratio; The ship model is obtained by processing according to the aforementioned ship model line drawing.

3. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 1, characterized in that, A test database is constructed based on the shallow water test results, the deep water test results, and the transition zone test results, specifically including: The results of the shallow water test, the deep water test, and the transition zone test are combined to obtain the comprehensive test results. Based on engineering experience, the results of the comprehensive test that do not meet the performance requirements are modified, and the modified comprehensive test results are determined as the test database.

4. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 1, characterized in that, Based on the experimental database and the simulation database, a virtual-agent model of the river-sea direct vessel is constructed, specifically including: The experimental database and the simulation database are merged to obtain a fused database; The fused database is preprocessed to obtain the training set; The ship design parameters of the ship model are used as input, and the corresponding performance test results in the training set are used as output to train the machine learning model, thereby obtaining the virtual-agent model of the river-sea direct vessel; each virtual-agent model corresponds to the prediction of a performance index.

5. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 1, characterized in that, Based on the aforementioned virtual-agent model, and with the goal of optimizing multiple performance indicators, an optimization algorithm is used to optimize the hull design parameters of the river-sea direct vessel, resulting in the optimal hull design parameters, specifically including: Construct the fitness function of the genetic optimization algorithm based on the virtual-proxy model; With the fitness function as the objective, the hull design parameters of the river-sea direct vessel are iteratively optimized multiple times to obtain the optimal hull design parameters of the river-sea direct vessel.

6. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 5, characterized in that, The expression for the fitness function is: ； in, Indicates ship design parameters The corresponding fitness function value; This represents the first performance metric value output by the first virtual-agent model. This represents the second performance metric value output by the second virtual-agent model; Indicates the first The output of the virtual-agent model is the first Various performance index values; This represents the weighting coefficient corresponding to the first performance index value; This represents the weighting coefficient corresponding to the second performance index value; Indicates the first The weighting coefficients corresponding to the performance index values.

7. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 4, characterized in that, The machine learning model is a support vector machine.

8. The composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels according to claim 1, characterized in that, The ship design parameters include: length, beam, draft, and bow and stern shape.

9. A composite environmental physical simulation system for optimizing the hull form of river-sea direct vessels, characterized in that, The composite environmental physical simulation system for optimizing the hull form of river-sea direct vessels adopts the composite environmental physical simulation method for optimizing the hull form of river-sea direct vessels as described in any one of claims 1-8. The composite environmental physical simulation system for optimizing the hull form of river-sea direct vessels includes: The test database acquisition module is used to acquire the test database of the river-sea direct vessel; the test database includes: test results obtained from the performance testing of the ship in the water tank based on the ship model and using a composite physical environment simulation method; the performance testing of the ship in the water tank includes: shallow water test of the ship model, deep water test of the ship model, and shallow-deep transition zone test of the ship model. The simulation database construction module is used to perform simulation parameterization calculations on the river-sea direct vessel and establish a simulation database; the experimental database includes: simulation results obtained by simulation performance testing based on the ship model using simulation methods. The virtual-proxy model construction module is used to construct a virtual-proxy model of the river-sea direct vessel based on the experimental database and the simulation database; the virtual-proxy model is used to predict various performance indicators of river-sea direct vessels with different ship type design parameters; The multi-objective optimization module is used to optimize the ship design parameters of the river-sea direct vessel based on the virtual-agent model, with the goal of optimizing multiple performance indicators, and to obtain the optimal ship design parameters of the river-sea direct vessel.

Images