A propeller-rudder integrated design optimization method and system introducing flow field coupling effect

Abstract

The embodiment of the application provides a ship propeller rudder integrated design optimization method and system introducing flow field coupling effect, and belongs to the field of ship engineering. The scheme uniformly models the parameters of the ship body, the propeller and the rudder surface, generates an integrated geometric model through surface fitting, carries out ship propeller rudder flow field coupling calculation based on the geometric model, and realizes nonlinear regression prediction of the design variables and the multi-working condition performance indicators through the construction of a proxy model; multi-optimization objectives are determined in combination with the actual sailing working condition characteristics of the ship, and a multi-objective optimization function is built by fusing various types of operation boundary constraints; relying on the multi-objective optimization function, the proxy model and the predicted performance indicators, a multi-objective optimization algorithm is used to globally optimize in the design variable space, and the optimized design parameters are output; a dynamic feedback closed loop mechanism is built to correct each parameter in real time in the iterative optimization process and continuously compensate for the prediction deviation caused by the flow field coupling. The application takes into account the multi-working condition operation constraints and can obtain an integrated optimal matching scheme suitable for the operation of the actual ship.

Description

Technical Field

[0001] This application relates to the field of marine engineering technology, and in particular to a method and system for optimizing the integrated design of ship propeller and rudder by introducing flow field coupling effects. Background Technology

[0002] Ships have varying requirements for propulsion efficiency, maneuverability, seakeeping performance, and energy conservation under different navigation conditions. Existing ship-propeller-rudder designs are mostly based on individual component optimization, meaning that the hull, propeller, and rudder surfaces are designed and evaluated independently. This individual optimization approach has its shortcomings. For example, the wake of the hull is highly coupled with the propeller-induced effect and the propeller-rudder interference effect. Existing methods often focus on optimizing a single component, making it difficult to reflect the true interference patterns of the integrated ship-propeller-rudder flow field, leading to inaccurate drag estimations and insufficient propulsion efficiency. Existing design processes generally rely on empirical formulas or low-precision calculation methods, making it difficult to handle complex optimization problems with multiple objectives and constraints. In particular, there is a lack of systematic methods to balance drag reduction, propulsion efficiency improvement, seakeeping performance enhancement, maneuverability enhancement, and rudder effectiveness improvement. The requirements for ship flow field characteristics, propulsion efficiency, and maneuverability vary under different waters and navigation conditions. Traditional optimization designs lack cross-condition adaptability, which can easily lead to decreased propulsion efficiency, increased drag, or enhanced local pressure pulsations under certain conditions.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this application is to propose a method and system for optimizing the integrated design of ship propeller and rudder by introducing flow field coupling effects, so as to achieve a comprehensive improvement in energy saving, consumption reduction and navigation performance.

[0005] To achieve the above objectives, one aspect of this application proposes a ship propeller-rudder integrated design optimization method that incorporates flow field coupling effects. The ship propeller-rudder integrated design optimization method includes: The hull parameters, propeller parameters, and rudder surface parameters are modeled in a unified manner, and a geometric model integrating the ship, propeller, and rudder is obtained through surface fitting. Based on the geometric model, flow field coupling calculations are performed on the hull, propeller and rudder, and nonlinear regression predictions are performed on the performance indicators under different combinations of design variables by constructing a surrogate model to obtain the performance indicators corresponding to each working condition. The optimization objective is determined based on the actual operating characteristics of the ship under different operating conditions, and a multi-objective optimization function is constructed in combination with the operating condition constraints. Based on the multi-objective optimization function, the surrogate model, and the performance index, the optimal design parameters are obtained by performing global optimization in the design variable space based on the optimization algorithm. The design parameters are corrected in real time during the optimization iteration process based on the dynamic feedback mechanism. The design parameters include design variables, model parameters, optimization parameters and operating condition constraint parameters.

[0006] In some embodiments, the design variables include stern streamline control point parameters, waterline parameters, pitch ratio, aspect ratio, blade tip thickness, rudder area coefficient, rudder longitudinal installation position, and rudder inclination angle; the performance indicators include hull drag coefficient, propeller induced drag, rudder additional drag, propulsion efficiency, seakeeping performance evaluation indicators, maneuverability evaluation indicators, and local pressure pulsation.

[0007] In some embodiments, flow field coupling calculations are performed on the hull, propeller, and rudder based on the geometric model, and nonlinear regression predictions are performed on performance indicators under different combinations of design variables by constructing a surrogate model to obtain performance indicators corresponding to each operating condition, including: The design variables are determined based on the parameters and a design variable space is constructed. Training samples are then generated within the design variable space. Based on the training samples, flow field coupling calculations were performed on the hull, propeller and rudder to obtain the corresponding performance indicators under different operating conditions. Establish a proxy model between the design variables and the performance indicators, and train the proxy model based on the training samples to obtain the trained proxy model; Based on the trained surrogate model, nonlinear regression prediction is performed on the performance indicators to obtain the predicted values ​​of the performance indicators for each working condition.

[0008] In some embodiments, flow field coupling calculations are performed on the hull, propeller, and rudder surface based on the training samples to obtain performance indicators corresponding to different operating conditions, including: A unified computational domain is established based on the actual installation positions and relative spatial relationships of the hull, propeller, and rudder. Within the unified computational domain, the inner-layer physical flow field coupling calculation is performed on the flow field around the hull, the rotating flow field of the propeller, and the flow field around the rudder surface to obtain the coupling calculation results. The flow field characteristics and performance indicators are extracted from the coupling calculation results, and the flow field characteristics and performance indicators are compared with the preset target values ​​or constraint thresholds under the corresponding navigation conditions to obtain the matching deviation. The design variables are iteratively adjusted based on the matching deviation to regenerate the integrated geometric model of the ship, propeller, and rudder. The integrated geometric model is recalculated for flow field coupling until the performance indicators corresponding to each working condition under steady-state conditions are obtained.

[0009] In some embodiments, the flow field data transfer process within the unified computing domain includes: Determine the static flow field region, the rotating flow field region, and the region interface within the unified computational domain; Computational grids are generated for the static flow field region and the rotating flow field region respectively, and donor grid cells and acceptor grid cells are determined according to the spatial positional relationship between the grid cells and the interface of the region. The interpolation weights are determined based on the donor grid cells and the recipient grid cells; Based on the propeller speed and the position of the grid points at the region interface relative to the propeller axis, the velocity coordinates of the static flow field region and the rotating flow field region are transformed. Based on the velocity coordinate transformation and the interpolation weight, the flow field parameters are transferred within the unified computational domain. The flow field data includes velocity, pressure, turbulence parameters, and flux.

[0010] In some embodiments, the process of determining steady-state conditions during flow field coupling calculations includes: By comprehensively using the residuals of the governing equations, the mass conservation error of the regional interface, and hydrodynamic monitoring data, the stability state of the inner physical flow field coupling is determined, and the first determination result is obtained. After the coupling calculation of the inner physical flow field reaches a stable state, the propulsion balance relationship between the propeller thrust and the total resistance generated by the hull, rudder surface and other appendages is judged to obtain the second judgment result. When both the first and second determination results are stable, the determination reaches the steady-state condition. When the steady-state condition is reached, the dynamic matching state of the geometric model and the flow field coupling calculation results is determined, and the feedback adjustment of the current design scheme is stopped when the dynamic matching state is reached.

[0011] In some embodiments, the operating conditions include shallow water navigation conditions, confined water navigation conditions, and deep water wave navigation conditions; the constraints include wake uniformity constraints, stability constraints, structural constraints, and draft and operating condition constraints.

[0012] In some embodiments, the optimization algorithm is the NSGA-III algorithm. Based on the multi-objective optimization function and the surrogate model, the NSGA-III algorithm is used to perform global optimization in the design variable space to obtain preferred design parameters, including: The population size is set according to the ship design variables and their range, and an initial population is randomly generated. Based on the initial population, the proxy model is invoked to perform nonlinear regression prediction on the performance indicators corresponding to each combination of design variables to obtain fitness evaluation results. Key individuals are selected from the initial population and ship propeller-rudder flow field coupling calculations are performed. The fitness evaluation results are corrected based on the actual calculated performance index values ​​to obtain the corrected evaluation results. Based on the correction and evaluation results, and considering the compliance of the constraints, all individuals in the initial population are sorted in a non-dominated manner to obtain the hierarchical sorted population sequence. Based on the objective dimension of the multi-objective optimization function, a uniformly distributed set of reference vectors is generated in the objective space to obtain a reference vector dataset. Based on the population sequence and the reference vector dataset, the spatial distance between each individual and the reference vector in the same non-dominated level is calculated. Based on the spatial distance, individuals in the same level are screened again to obtain a set of high-quality individuals. The set of high-quality individuals is iteratively sorted based on genetic operators, and the convergence status is determined by convergence criteria to output the optimal design parameters.

[0013] To achieve the above objectives, another aspect of this application proposes a ship propeller-rudder integrated design optimization system that incorporates flow field coupling effects. The ship propeller-rudder integrated design optimization system includes: The ship parametric modeling module is used to uniformly model the hull parameters, propeller parameters, and rudder parameters, and obtain an integrated geometric model of the ship, propeller, and rudder through surface fitting. The flow field coupling calculation module is used to perform flow field coupling calculations on the hull, propeller and rudder based on the geometric model, and obtain the performance index data corresponding to each working condition. The operating condition constraint management module is used to transform the ship's operating characteristics under different navigation conditions into design constraints and to map complex operating environments into design boundary conditions, thereby obtaining the constraint conditions for different operating conditions. The rapid optimization and optimization module is used to perform nonlinear regression prediction of performance indicators under different combinations of design variables by constructing a surrogate model based on performance index data and constraints, and to perform global optimization in the design variable space based on a multi-objective optimization algorithm to obtain the optimal design parameters. The dynamic feedback module, based on the dynamic feedback mechanism, corrects the design parameters in real time during the optimization iteration process. The design parameters include design variables, model parameters, optimization parameters, and operating condition constraint parameters.

[0014] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0016] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0017] The embodiments of this application include at least the following beneficial effects: This application provides a method and system for integrated design optimization of ship propeller and rudder by introducing flow field coupling effects. This scheme models the hull, propeller, and rudder surfaces with unified parameters and generates an integrated geometric model through surface fitting. By unifying the parameters of the three types of components into the same modeling system, it can eliminate the geometric connection discontinuities and flow field interface mismatches caused by separate component modeling, and completely restore the spatial assembly position, external curved surface, and true geometric boundaries of mutual interference between the ship, propeller, and rudder. Based on this geometric model, flow field coupling calculations of ship, propeller, and rudder can accurately capture the mutual flow field coupling effects between the hull wake, propeller wake vortex, and rudder surface disturbance, and obtain high-precision sample performance data. By constructing a surrogate model, nonlinear regression prediction of design variables and multi-condition performance indicators can be realized. Low-cost and fast regression prediction replaces full-volume and time-consuming CFD simulation, significantly reducing the computing cost of performance calculation under multiple variables and multiple conditions, and quickly outputting performance indicators corresponding to any combination of design variables in batches.

[0018] By combining the actual navigation conditions of the ship, multiple optimization objectives are determined, and a multi-objective optimization function is constructed by integrating various operational boundary constraints. Abandoning the limitations of single-performance index optimization, and aligning with the actual operational needs of the ship under all navigation conditions, multiple optimization objectives are simultaneously determined. At the same time, ship geometric constraints, structural strength, stability, draft, and operational limitations under different speeds / load conditions are transformed into standardized constraints. This constructs a multi-objective optimization function that considers multiple performance requirements and fits the actual ship operational boundaries, ensuring that the subsequent optimization process is always carried out within a feasible design space that meets the actual usage requirements of the ship, avoiding optimization results that deviate from engineering reality. Based on the multi-objective optimization function, surrogate model, and predicted performance indicators, a multi-objective optimization algorithm is used to globally optimize in the design variable space, outputting optimal design parameters. By traversing a massive number of design parameter combinations, the Pareto optimal solution set that considers multiple conflicting performance objectives is automatically selected, escaping local optima and extensively exploring ship-propeller-rudder matching schemes with superior overall performance within the design variable space, achieving integrated initial selection of schemes under multiple variables, constraints, and objectives.

[0019] A dynamic feedback closed-loop mechanism is established to correct various parameters in real time during the iterative optimization process, continuously compensating for prediction deviations caused by flow field coupling. By correcting relevant parameters of the geometric model, the prediction errors between the surrogate model and the actual CFD calculation are compensated. By adjusting optimization parameters such as the population size and number of iterations of the optimization algorithm, the global search balance is improved. By dynamically fine-tuning the constraint thresholds of various navigation conditions, the performance boundary changes caused by flow field coupling are adapted. The prediction deviation of the surrogate model, the search defects of the optimization algorithm, and the mismatch of the operating condition constraints are continuously weakened. The optimal parameters are continuously refined through iteration, and finally, a ship-propeller-rudder integrated design scheme with higher flow field coupling effect matching, stronger adaptability to all operating conditions, and better engineering feasibility is obtained. Attached Figure Description

[0020] Figure 1 This is a schematic diagram illustrating the steps of an integrated ship propeller and rudder design optimization method that incorporates flow field coupling effects, as provided in an embodiment of this application. Figure 2 This is a flowchart of a ship propeller-rudder integrated design optimization method that introduces flow field coupling effect, provided in an embodiment of this application. Figure 3 This is a schematic diagram of a ship propeller-rudder integrated design optimization system that incorporates flow field coupling effects, as provided in an embodiment of this application. Figure 4 This is a detailed structural block diagram of a ship propeller-rudder integrated design optimization system that introduces flow field coupling effect, provided in an embodiment of this application. Figure 5 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0021] 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 of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0022] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0023] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0025] To address the shortcomings of existing technologies that primarily focus on individual component optimization in ship-propeller-rudder design, which suffers from insufficient flow field coupling, limited optimization methods, and inadequate adaptability and robustness, this application provides a ship-propeller-rudder integrated design optimization method and system that incorporates flow field coupling effects. This method simultaneously considers hull wake distribution, propeller-induced effects, and propeller-rudder interference effects to achieve integrated ship-propeller-rudder optimization, overcoming the drag prediction bias and insufficient propulsion efficiency issues caused by existing single-component optimization. Furthermore, it constructs a joint framework based on surrogate models and multi-objective optimization algorithms, breaking through the limitations of traditional empirical formulas and low-precision methods to achieve multi-objective collaborative optimization that reduces hull drag, improves propulsion efficiency, enhances seakeeping performance, strengthens maneuverability, and improves rudder effectiveness. Under different navigation conditions, an optimization process with a dynamic feedback mechanism is established to improve the adaptability and robustness of the design method to complex conditions and constraints.

[0026] This application provides a method and system for optimizing the integrated design of a ship's propeller and rudder by incorporating flow field coupling effects, relating to the field of marine engineering. This method and system can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited thereto. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the method and system for optimizing the integrated design of a ship's propeller and rudder by incorporating flow field coupling effects, but is not limited to the above forms.

[0027] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0028] Please see Figure 1 and Figure 2 As shown, Figure 1 This is a schematic diagram of an optional step in a ship propeller-rudder integrated design optimization method that incorporates flow field coupling effects, as provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S1 to S5: S1, the hull parameters, propeller parameters and rudder surface parameters are modeled in a unified manner, and the geometric model of the ship, propeller and rudder integrated is obtained by surface fitting; In some embodiments, hull parameters include length L, beam B, draft T, stern streamline control point, and waterline parameters; propeller parameters include number of blades Z, pitch ratio P / D, and aspect ratio. and leaf tip thickness ; Control surface parameters include control surface area coefficient Longitudinal mounting position of the rudder and rudder tilt angle .

[0029] Hull parameters: Waterline, transverse section, longitudinal section, and stern streamlines were extracted based on the baseline hull form and fitted using B-spline curves. The hull geometry was parametrically controlled by adjusting the main dimensions: length L, beam B, draft T, stern streamline control points, and waterline shape parameters. Specifically, the main dimensions (length L, beam B, and draft T) control the longitudinal, transverse, and vertical dimensions of the hull, respectively; the stern streamline control points adjust the degree of stern contraction, bottom rise, and streamline orientation; and the waterline shape parameters adjust the waterline curvature and stern contraction shape. After parameter changes, the hull surface was refitted based on the updated waterline, transverse section, longitudinal section, and stern streamlines, ensuring continuity and smoothness between adjacent curves and surfaces.

[0030] Propeller parameters: Based on blade profiles at different radii of a reference propeller, using the number of blades Z, pitch ratio P / D, and aspect ratio. and leaf tip thickness The propeller geometry is parametrically controlled. Specifically, the number of blades Z controls the number and circumferential distribution of blades; the pitch ratio P / D adjusts the pitch distribution at various radii; and the aspect ratio... Used to adjust the chord length and area distribution of the blade; leaf tip thickness Used to adjust the thickness of the blade tip region. After parameter changes, the contour and spatial position of each blade profile are redefined, and a new propeller blade surface is generated through surface fitting.

[0031] Control surface parameters: Based on the reference control surface profile and control surface profile, the control surface area coefficient is adopted. longitudinal mounting position of the rudder and rudder tilt angle Parametric control of the control surfaces is implemented. Among these parameters is the control surface area coefficient. Used to adjust the chord length, span, or a combination of both of the rudder surface; longitudinal mounting position of the rudder. Used to control the position of the rudder surface along the length of the ship; rudder inclination angle Used to control the rotation angle of the control surface around the mounting axis. After the parameters change, the control surface profile, control surface section position and mounting attitude are redefined, and a new control surface model is generated.

[0032] A geometric model integrating the ship, propeller, and rudder is generated based on the fitted hull surface, propeller blade surface, and rudder surface models.

[0033] S2, based on the geometric model, performs flow field coupling calculations on the hull, propeller and rudder, and uses a proxy model to perform nonlinear regression prediction of performance indicators under different combinations of design variables to obtain the performance indicators corresponding to each working condition. In some embodiments, flow field coupling calculations are performed on the hull, propeller, and rudder based on a geometric model, and a surrogate model is constructed to perform nonlinear regression predictions on the performance indicators of the training samples under different combinations of design variables, thereby obtaining the performance indicators corresponding to each operating condition, including: S21, Determine the design variables based on the parameters and construct the design variable space, and generate training samples within the design variable space; S22, based on the training samples, perform flow field coupling calculations on the hull, propeller and rudder to obtain the corresponding performance indicators under different working conditions; In some embodiments, design variables include stern streamline control point parameters, waterline parameters, pitch ratio, aspect ratio, blade tip thickness, rudder area coefficient, rudder longitudinal installation position, and rudder inclination angle; performance indicators include hull drag coefficient, propeller induced drag, rudder additional drag, propulsion efficiency, seakeeping performance evaluation indicators, maneuverability evaluation indicators, and local pressure pulsation.

[0034] The structural parameters of the hull, propeller, and rudder are selected as design variables to construct the design variable space, which is represented as follows: Where: x represents the design variable vector, and d represents the dimension of the design variables. For example, in this embodiment... ;in, These are the streamline control point parameters for the stern, with values ​​ranging from 0.97 to 1.03. This is a waterline profile parameter, with a value range of 0.97 to 1.03; The pitch ratio P / D ranges from 0.88 to 1.04. for aspect ratio The value ranges from 0.55 to 0.70; Thickness at the leaf tip The value ranges from 10.8 to 13.2 mm; Rudder area coefficient The value ranges from 0.018 to 0.022; The longitudinal mounting position of the rudder The value ranges from 0.972 to 0.986; Rudder angle The value range is -1.5° to 1.5°.

[0035] Within the constructed design variable space, the Latin hypercube sampling method is used to generate initial training samples. During sample generation, the ship-propeller interference effect and the propeller-rudder interference coupling effect are considered. CFD calculations are used to obtain the total hull resistance, propeller efficiency, rudder surface load, propulsion efficiency, seakeeping performance, maneuverability, and flow field distribution under different combinations of design variables, and the corresponding performance indicators are obtained, expressed as follows: In the formula, Indicates the hull drag coefficient. Indicates the induced drag of the propeller. Indicates the additional drag of the rudder. Indicates the efficiency of advancement. Indicates the evaluation index of wave resistance performance. Indicates the performance evaluation index of handling. This indicates localized pressure pulsation.

[0036] For each set of design variables, the corresponding geometric models of the hull, propeller, and rudder are reconstructed, and the hull, propeller, and rudder are imported into a unified computational domain according to their actual installation positions. Within this unified computational domain, coupled flow field calculations are performed on the hull, propeller, and rudder. During the calculation process, the flow fields around the hull, the rotating flow field of the propeller, and the flow fields around the rudder are iteratively coupled and solved until the hull drag, propeller thrust and torque, and rudder load reach a stable state. This yields the ship-propeller-rudder coupled flow field under the interaction of the hull wake, propeller-induced flow, and rudder forces, resulting in performance indicators for different operating conditions.

[0037] S23, Establish a proxy model between design variables and performance indicators, and train the proxy model based on training samples to obtain the trained proxy model; S24. Based on the trained surrogate model, nonlinear regression prediction is performed on the performance indicators to obtain the predicted values ​​of the performance indicators for each working condition.

[0038] Based on the initial sample data, this embodiment uses an RBF network to perform nonlinear regression approximations on the total hull resistance, propulsion efficiency, rudder effectiveness, seakeeping performance, maneuverability, and flow field coupling indices, replacing some of the high-fidelity CFD calculations and improving optimization efficiency, as expressed below: ; In the formula, φ(r) represents the predicted performance index by the surrogate model; φ(r) is the radial basis function, which can be a Gaussian kernel. Or multiple secondary cores; The coefficients are undetermined and obtained by fitting the sample point data; P(x) is a low-order polynomial term used to ensure global approximation performance. Let N be the i-th design variable, and N be the number of design variables.

[0039] The parameters are solved by establishing a system of equations, expressed as: Kλ+P=Y; In the formula, K is the kernel matrix composed of radial basis functions; λ is the weight coefficient vector to be solved; P is a polynomial matrix; and Y is the sample performance index vector.

[0040] It should be noted that the embodiments of this application construct a surrogate model based on RBF radial basis function networks. First, the relevant design variables of the hull, propeller, and rudder are determined and a design variable space is constructed. Training samples are generated within this space, and integrated flow field coupling CFD calculations are performed on each set of samples to obtain sample data of various performance indicators under different operating conditions. Then, the mapping relationship between design variables and performance indicators is established, relying on formulas... An RBF network is constructed, and the model is trained by solving the weight coefficients using the equation system Kλ + P = Y. Nonlinear fitting is achieved using a Gaussian kernel or multiple quadratic kernels, and low-order polynomial terms ensure global approximation capability. After training, K-fold cross-validation is used to check the prediction error of the surrogate model. If the prediction error does not exceed a preset error threshold, the surrogate model is used in subsequent multi-objective optimization; if the prediction error exceeds the preset error threshold, the surrogate model parameters, kernel function parameters, or training sample data are corrected until the surrogate model meets the accuracy requirements. Finally, the qualified RBF surrogate model is used to achieve fast nonlinear regression prediction of various performance indicators such as total hull drag, propulsion efficiency, rudder effectiveness, and flow field coupling index, thereby replacing a large number of time-consuming CFD simulations and effectively reducing the computational cost in the iterative optimization process.

[0041] In some embodiments, flow field coupling calculations are performed on the hull, propeller, and rudder surface based on training samples to obtain performance indicators corresponding to different operating conditions, including: S221, a unified calculation domain is established according to the actual installation positions and relative spatial relationships of the hull, propeller and rudder. In some embodiments, the flow field data transfer process within the unified computing domain includes: Define the static flow field region, the rotating flow field region, and the region interface within the unified computational domain; The flow field coupling effect in this embodiment includes two interconnected levels: the first level is the physical flow field coupling between the flow field around the hull, the propeller rotation flow field, and the flow field around the rudder surface. That is, the wake of the hull stern affects the propeller inflow, the induced flow generated by the propeller affects the force on the rudder surface, and the flow field changes caused by the propeller and rudder surface in turn affect the flow field around the hull stern. The second level is the dynamic matching between parametric modeling and CFD (Computational Fluid Dynamics) flow field calculation. That is, based on the flow field characteristics and performance indicators obtained from CFD calculations, the design variables of the hull, propeller, and rudder surface are jointly adjusted, and an integrated geometric model is regenerated for the next coupling calculation. Thus, a ship-propeller-rudder dynamic matching process is formed, based on the inner physical flow field coupling calculation and with the outer parameter feedback adjustment as a closed loop.

[0042] For the integrated geometric model of the hull, propeller and rudder obtained by parametric modeling, a unified calculation domain of ship-propeller-rudder is established according to the actual installation position and relative spatial relationship of the hull, propeller and rudder. In this calculation domain, the area where the hull and rudder are located is set as a static flow field area, the area around the propeller is set as a rotating flow field area, and the interface between the static flow field area and the rotating flow field area is set.

[0043] Computational grids are generated for the static flow field region and the rotating flow field region respectively. The donor grid cells and acceptor grid cells are determined based on the spatial positional relationship between the grid cells and the region interface. The interpolation weights are determined based on the donor and recipient grid cells; Computational meshes are generated for the static flow field region and the rotating flow field region respectively. The interface between the two flow field regions can be set using a non-conformal mesh. The mesh nodes and mesh faces on both sides of the interface do not need to completely overlap. Instead, the donor mesh cells for providing flow field data and the receiver mesh cells for receiving flow field data are determined based on the spatial relationship between the mesh cells and the interface. In this embodiment, the interpolation weight is determined based on the overlap area between the receiver mesh face and the adjacent donor mesh face, and the velocity, pressure, and turbulence parameters in the donor mesh cells are transferred to the receiver mesh face according to the interpolation weight. Turbulence parameters include turbulent kinetic energy and turbulent dissipation rate or specific dissipation rate. In other embodiments, the interpolation weight can also be determined based on the distance or projected area between the donor mesh cell and the receiver mesh face.

[0044] Pressure and turbulence parameters are interpolated based on parameter values ​​in adjacent grid cells on both sides of the interface and corrected during the pressure-velocity coupled solution process. Simultaneously, mass flux and momentum flux are calculated based on fluid density, normal velocity at the interface, and grid surface area. The interpolated fluxes are then conserved to maintain overall mass flow rate balance on both sides of the interface and keep momentum transfer error within a preset allowable range. This avoids false mass sources or significant momentum losses at the interface caused by non-conformal grid interpolation. In this embodiment, when the relative deviation of mass flow rate between the static flow field side and the rotating flow field side does not exceed 0.2%, the data transfer at the interface is deemed to meet the preset mass conservation requirement.

[0045] Based on the propeller speed and the position of the grid points at the region interface relative to the propeller axis, velocity coordinate transformation is performed between the static flow field region and the rotating flow field region. Since the static and rotating flow fields use stationary and rotating coordinate systems respectively, the velocity transfer across the interface requires coordinate transformation based on the propeller speed and the position of the grid points at the interface relative to the propeller axis. Specifically, when the velocity is transferred from the static to the rotating flow field, the absolute velocity in the stationary coordinate system is converted to the relative velocity in the rotating coordinate system; conversely, when the velocity is transferred from the rotating to the static flow field, the relative velocity in the rotating coordinate system is converted to the absolute velocity in the stationary coordinate system.

[0046] Based on velocity coordinate transformation and interpolation weights, the flow field parameters are transferred within a unified computational domain. The flow field data includes velocity, pressure, turbulence parameters, and flux.

[0047] For steady-state calculations in rotating coordinate systems, the grid positions of the rotating flow field region remain unchanged. Flow field data transfer between the stationary and rotating flow field regions is achieved through velocity coordinate transformation, flow field parameter interpolation, and flux conservation correction. For transient calculations using slip grids, the rotating flow field region rotates with the propeller. At each time step, based on the relative positions of the grids at the interface between the rotating and stationary flow field regions, the donor and recipient grid elements and their corresponding interpolation weights are redefined, thereby updating the velocity, pressure, turbulence parameters, and flux on both sides of the interface.

[0048] S222, within a unified computational domain, performs coupled calculations of the inner-layer physical flow field of the flow field around the hull, the rotating flow field of the propeller, and the flow field around the rudder surface, and obtains the coupled calculation results. Under a given propeller speed, the inner-layer physical flow field coupling calculation is performed on the flow field around the hull, the rotating flow field of the propeller, and the flow field around the control surface. In each coupling iteration step, the continuity equation, momentum equation, and turbulence equation in the static flow field region and the rotating flow field region are solved alternately, and the flow velocity, pressure, turbulence parameters, mass flux, and momentum flux are repeatedly updated through the region interface.

[0049] In the current coupled iteration, the static flow field region where the hull and rudder are located is first solved based on the existing flow field parameters, obtaining the velocity, pressure, and turbulence parameter distributions at the stern of the hull, in front of the propeller disk, and near the rudder. The non-uniform wake formed at the stern of the hull is transmitted to the rotating flow field region where the propeller is located via the region interface, serving as the actual inflow condition for the propeller. Due to the different inflow velocities and directions at different radial and circumferential positions on the propeller disk, the non-uniform wake at the stern of the hull causes changes in the pressure distribution, disk load, thrust, and torque on the propeller blade surface. Under the non-uniform inflow condition, the rotating flow field of the propeller is solved, obtaining the pressure distribution, thrust, torque, disk load on the propeller blades, as well as the axial acceleration flow and circumferential rotation-induced flow generated by the propeller rotation. The velocity, pressure, and turbulence parameters at the propeller outlet are transmitted back to the static flow field region after coordinate transformation and region interface interpolation, and act on the rudder and stern of the hull located downstream of the propeller. The flow field around the rudder surface is resolved based on the updated propeller-induced flow to obtain the rudder surface pressure distribution, normal force, lateral force, load, and additional drag. Simultaneously, the axial acceleration flow generated by the propeller, the circumferential rotational induced flow, and the flow field changes caused by the forces on the rudder surface alter the velocity, pressure, and turbulence fields near the stern. These flow field changes are then re-involved in the solution of the stern flow field, causing alterations in the stern surface pressure distribution, hull drag, and propeller wake distribution. The updated propeller wake then serves as the actual propeller inflow condition in the next coupled iteration.

[0050] Therefore, the influence of the stern wake on the propeller load, the influence of the propeller-induced flow on the rudder surface force, and the reaction of the propeller and rudder surface on the stern flow field are all achieved through the bidirectional transmission and repeated updating of the flow field parameters and flux at the regional interface, thus forming a closed-loop physical flow field coupling relationship between the stern wake, the propeller-induced flow, and the rudder surface force.

[0051] To prevent excessive changes in flow field parameters between adjacent coupling iterations, relaxation processing can be applied to the velocity, pressure correction, turbulence parameters, and interface flux after the transfer from the regional interface. This is achieved by combining the current calculated value with the value from the previous iteration according to a preset ratio, and using the combined result as the input for the next coupling iteration. In this embodiment, the relaxation ratio can be set to 0.3–0.7. The above coupling process repeatedly executes the following steps: solving the flow field at the stern of the hull, transferring the flow from the stern of the hull to the rotating flow field region, solving the rotating flow field of the propeller, transferring the propeller-induced flow back to the stationary flow field region, and updating the flow field at the rudder and stern of the hull, until the residuals of the governing equations, the conservation error of the regional interface, and the hydrodynamic monitoring quantities of the hull, propeller, and rudder simultaneously meet the preset stability conditions.

[0052] S223, extract flow field characteristics and performance indicators from the coupled calculation results, and compare the flow field characteristics and performance indicators with the preset target values ​​or constraint thresholds under the corresponding navigation conditions to obtain the matching deviation; After completing the inner-layer physical flow field coupling calculation of the current propeller-rudder geometry model, flow field characteristics and performance indicators reflecting the matching degree of the hull, propeller, and rudder surface are extracted from the CFD calculation results. These flow field characteristics include the axial velocity distribution on the propeller disk, the circumferential velocity distribution, the non-uniformity of the propeller wake, the load distribution on the propeller disk, the direction of the propeller-induced flow, the actual inflow angle of the rudder surface, and the pressure distribution on the rudder surface. Performance indicators include the hull drag coefficient. Propeller-induced drag Additional drag from the rudder , Promotion efficiency Wave resistance performance evaluation indicators Handling performance evaluation indicators and local pressure pulsation .

[0053] By comparing the flow field characteristics and performance indicators with preset target values ​​or constraint thresholds under the corresponding navigation conditions, the matching deviation between the current geometric parameters of the hull, propeller, and control surfaces is determined. In this embodiment, the matching deviation includes the deviation between the wake distribution at the stern of the hull and the inflow requirements of the propeller disk, the deviation between the propeller thrust and torque and the ship's resistance, the deviation between the direction of the propeller induced flow and the control surface installation state, and the deviation between the hull resistance, propulsion efficiency, control surface load, seakeeping performance, maneuverability, local pressure pulsation, and the corresponding target values.

[0054] S224, Iteratively adjust the design variables based on the matching deviation, and regenerate the integrated geometric model of the ship, propeller and rudder; S225, the flow field coupling calculation is re-performed on the integrated geometric model until the performance indicators corresponding to each working condition under steady-state conditions are obtained.

[0055] When the non-uniformity of the propeller wake exceeds a preset threshold, or when the local low-speed zone and flow direction deviation zone on the propeller disk do not meet preset requirements, the streamline control point parameters or waterline parameters at the stern are adjusted to change the degree of stern contraction, pressure recovery process, and propeller wake distribution. When the propeller thrust, torque, disk load, or propulsion efficiency do not meet preset requirements, the pitch ratio (P / D) and aspect ratio are adjusted. Or the thickness of the leaf tip Adjustments are made to alter the propeller blade load level, thrust-torque matching, and backflow characteristics. When the actual inflow angle, pressure distribution, load, maneuverability evaluation indicators, or additional drag of the rudder do not meet preset requirements, the rudder area coefficient is adjusted. Longitudinal mounting position of the rudder or rudder tilt angle Adjustments are made to change the position of the control surface relative to the propeller-induced flow, the effective flow-receiving area of ​​the control surface, and the flow-receiving direction of the control surface.

[0056] The adjustment amounts of the aforementioned design variables are determined based on the RBF surrogate model established using existing design samples and CFD calculation results. The RBF surrogate model is used to predict the flow field characteristics and performance indicators corresponding to different combinations of design variables. Combined with multi-objective optimization results, and under the conditions of satisfying wake uniformity constraints, stability constraints, structural constraints, and draft and operating condition constraints, it determines the combination of design variables to be used in the next parametric modeling iteration. In each dynamic matching iteration, the parametric modeling module generates an integrated ship-propeller-rudder geometric model based on the current design variables. The CFD calculation module performs physical flow field coupling calculations on the geometric model. The dynamic feedback module determines the adjustment direction and magnitude of the design variables based on the matching deviation between the calculated flow field characteristics and performance indicators and the preset target values ​​or constraint thresholds.

[0057] After adjusting the design variables, the parametric modeling module regenerates the integrated geometric model of the hull, propeller, and rudder, and updates the relative installation relationships and corresponding computational meshes among them. Subsequently, it performs another round of regional interface data transfer and ship-propeller-rudder physical flow field coupling calculation. This forms an outer dynamic matching closed loop: parametric modeling, CFD physical flow field coupling calculation, flow field feature and performance index extraction, matching deviation judgment, design variable adjustment, and geometric model reconstruction. The inner physical flow field coupling is used to obtain the flow field and hydrodynamic results under the interaction of the hull, propeller, and rudder. The outer dynamic matching is used to feed back the flow field and hydrodynamic results to the parametric modeling module and adjust the geometric parameters. The adjustment of the outer geometric parameters further changes the geometric boundaries and flow field conditions for the next inner flow field coupling calculation.

[0058] When the steady-state condition is reached, i.e., the inner-layer physical flow field coupling calculation reaches a stable state and the propulsion balance judgment is completed, the pressure and viscous forces on the hull, propeller blades, and rudder surfaces are integrated to obtain the hull drag coefficient. Propeller-induced drag and additional drag of the rudder The propulsion efficiency is obtained based on the propeller thrust, torque, speed, and input power. Among them, propeller-induced drag This represents the increase in drag of the stern and appendages under the influence of propeller-induced flow relative to the state without propeller action, or the induced drag component obtained from ship-propeller-rudder coupling calculations according to a preset drag decomposition method. In confined water navigation conditions, maneuverability evaluation indicators are obtained based on rudder surface normal force, rudder surface lateral force, rudder surface load, and ship bow moment. Under deep-water wave navigation conditions, seakeeping performance evaluation indicators are obtained based on the ship's pitch response, heave response, and bow vertical acceleration. Local pressure pulsations are obtained based on pressure changes at preset monitoring points on the stern of the hull, propeller blades, and rudder surfaces. .

[0059] Thus, the hull resistance coefficients under various navigation conditions are obtained. Propeller-induced drag Rudder additional drag , Promotion efficiency Wave resistance performance evaluation indicators Handling performance evaluation indicators and local pressure pulsation The performance metrics were then used for training the RBF surrogate model and subsequent multi-objective optimization.

[0060] By employing coordinate system transformation, flow field parameter interpolation, and flux conservation correction at the regional interface, the transfer of velocity, pressure, and turbulence parameters between the static and rotating flow field regions is achieved. Through bidirectional transfer and repeated updates between the stern flow field, the rotating propeller flow field, and the flow field around the rudder, physical flow field coupling between the hull wake, propeller-induced flow, and rudder forces is realized. By feeding back the flow field characteristics and performance indicators obtained from CFD coupling calculations to the parametric modeling module, the design variables of the hull, propeller, and rudder are jointly adjusted, achieving dynamic matching between the parametric geometric model and the CFD flow field calculation results. Thus, this embodiment forms a closed-loop coupling process where changes in geometric parameters cause changes in the flow field, changes in the flow field generate performance deviations, and performance deviations are used to adjust the geometric parameters in reverse, achieving coordinated matching between the geometry of the hull, propeller, and rudder and the flow field performance.

[0061] In some embodiments, the process of determining steady-state conditions during flow field coupling calculations includes: By comprehensively using the residuals of the governing equations, the mass conservation error of the regional interface, and hydrodynamic monitoring data, the stability state of the inner physical flow field coupling is determined, and the first determination result is obtained. In this embodiment, when the normalized residuals of the continuity equation, momentum equation, and turbulence equation are all no greater than... When the relative deviation of mass flow rate on both sides of the interface is no greater than 0.2%, and the relative change rate of hull resistance, propeller thrust, propeller torque, and rudder load within the continuous monitoring interval is no greater than 0.5%, the inner physical flow field coupling calculation under the current propeller speed condition is considered to have reached a stable state. When judging hydrodynamic monitoring quantities, the 500 consecutive iterations can be divided into two adjacent monitoring intervals. The average values ​​of hull resistance, propeller thrust, propeller torque, and rudder load within each monitoring interval are calculated. When the change in the average value between adjacent monitoring intervals does not exceed 0.5%, the corresponding hydrodynamic monitoring quantity is considered to have reached a stable state.

[0062] For transient calculations using the sliding mesh method, the stable flow field obtained from steady-state calculations can be used as the initial flow field. Then, the rotating flow field region is made to rotate with the propeller according to a preset time step. In this embodiment, each time step corresponds to a 2° propeller rotation, and at least 6 propeller rotation cycles are calculated continuously. The calculation results of the last 3 rotation cycles are used to determine the periodic stability. When the periodic average changes of hull resistance, propeller thrust, propeller torque, and rudder load do not exceed 1% within 3 consecutive propeller rotation cycles, and the fluctuation amplitude and phase of the corresponding hydrodynamic monitoring quantities show repeated periodic changes, the transient physical flow field coupling calculation is determined to have reached a periodic stable state.

[0063] After the coupling calculation of the inner physical flow field reaches a stable state, the propulsion balance relationship between the propeller thrust and the total resistance generated by the hull, rudder surface and other appendages is judged to obtain the second judgment result. For steady-state straight-line navigation, propulsion balance is considered complete when the relative difference between propeller thrust and total ship resistance does not exceed 2%. When the relative difference exceeds 2%, the propeller speed is adjusted, and the inner-layer physical flow field coupling calculation is re-performed under the adjusted propeller speed conditions. For unsteady conditions such as wave navigation or maneuvering, propulsion balance is determined using the average thrust and average total resistance within a preset statistical period. When the relative difference between the average thrust and average total resistance does not exceed a preset threshold, the corresponding unsteady condition is considered to have reached a statistically significant propulsion balance state.

[0064] When both the first and second determination results are stable, the determination reaches the steady-state condition. When the steady-state condition is reached, the dynamic matching state of the geometric model and the flow field coupling calculation results is determined, and the feedback adjustment of the current design scheme is stopped when the dynamic matching state is reached.

[0065] For the outer layer parametric modeling and CFD dynamic matching process, when the propeller wake nonuniformity, propeller thrust and torque, rudder surface load, as well as hull resistance, propulsion efficiency, seakeeping performance, maneuverability and local pressure pulsation all meet the corresponding objectives or constraints, and the changes in design variables and performance indicators in two adjacent dynamic matching iterations do not exceed the preset threshold, it is determined that the parametric geometric model and the CFD flow field calculation results have reached a dynamic matching state, and the feedback adjustment of the current design scheme is stopped.

[0066] S3. Determine the optimization objective based on the actual operating characteristics of the ship under different operating conditions, and construct a multi-objective optimization function in combination with the operating condition constraints. In some embodiments, the operating conditions include shallow water navigation conditions, confined water navigation conditions, and deep water wave navigation conditions; the constraints include wake uniformity constraints. Stability constraints Structural constraints and draft and operating condition constraints .

[0067] Specifically, the optimization objectives are determined based on the actual operating characteristics of the ship under various working conditions. The first optimization objective is to minimize the weighted total resistance coefficient, expressed as: ; In the formula, The weighting factor is set according to different navigation conditions.

[0068] The second optimization objective is to maximize propulsion efficiency, expressed as: ; The third optimization objective is to maximize the wavekeeping performance, expressed as: ; The fourth optimization objective is to maximize maneuverability, expressed as: ; The fifth optimization objective is to minimize the pressure pulsation amplitude, expressed as: .

[0069] Then, a multi-objective optimization function is constructed by combining the operating constraints under various working conditions, including wake uniformity constraints. Used to constrain the non-uniformity of the velocity distribution in front of the impeller to not exceed a preset threshold; stability constraint Used to match the ship's static water stability and related seaworthiness requirements; structural constraints Used to constrain the stress on propeller blades and control surfaces from exceeding material limits; draft and operating condition constraints. This is used to ensure that the vessel meets operational conditions in various scenarios, including shallow water, confined water, and deep water with high waves. For example, in shallow water navigation, the non-uniformity of the flow velocity distribution in front of the propeller does not exceed a preset threshold; in confined water navigation, the maneuverability evaluation index... Not lower than the preset threshold; wave resistance performance evaluation indicators under deep-water wave navigation conditions. Local pressure pulsation not lower than a preset threshold The maximum stress on the blades and control surfaces does not exceed the preset upper limit; under all operating conditions, the maximum stress on the blades and control surfaces does not exceed the allowable stress of the corresponding materials. The final multi-objective optimization function is expressed as: ; In the formula, This represents the j-th constraint.

[0070] S4. Based on the multi-objective optimization function, surrogate model and performance indicators, global optimization is performed in the design variable space based on the optimization algorithm to obtain the optimal design parameters. In some embodiments, the optimization algorithm is the NSGA-III (Reference Point-Based Non-Dominated Genetics) algorithm. Based on the multi-objective optimization function and the surrogate model, the NSGA-III algorithm performs global optimization in the design variable space to obtain the preferred design parameters, including: S41, Set the population size according to the ship design variables and their value range, and randomly generate the initial population; The population size is pre-set, and the initial population individuals are randomly generated by combining the value range of all design variables of the hull, propeller, and rudder. Each population individual is composed of a complete combination of design variables jointly coded by hull parameters, propeller parameters, and rudder parameters. The range of design variables is jointly limited by wake uniformity constraints, structural constraints, stability constraints, and draft and operating condition constraints.

[0071] S42, Based on the initial population, the proxy model is invoked to perform nonlinear regression prediction on the performance indicators corresponding to each combination of design variables to obtain the fitness evaluation results; The flow field coupling proxy model constructed by S2 is called. The encoded design variables of each individual in the first generation population are used as input. The performance index corresponding to each combination of design variables is predicted by nonlinear regression. The multi-objective function value F(x) corresponding to all individuals is output, and the preliminary fitness assessment results of the population are obtained.

[0072] S43, key individuals are selected from the initial population and the ship propeller-rudder flow field coupling calculation is performed. The fitness evaluation results are corrected based on the actual calculated performance index values ​​to obtain the corrected evaluation results. Three key individuals were selected from the initial population: Pareto frontier elite individuals, individuals with high uncertainty in surrogate model predictions, and individuals at the constraint boundary criticality. High-precision CFD calculations were performed on the selected key individuals using a coupled ship-propeller-rudder flow field to obtain true high-precision performance index values. These calculated performance index values ​​were then used as new training samples and fed back to the surrogate model to correct its regression accuracy in real time, resulting in an updated high-precision surrogate model. Based on the updated high-precision surrogate model, nonlinear regression predictions were performed again on the performance indices corresponding to each design variable combination to obtain the corrected fitness evaluation results, i.e., the corrected evaluation results.

[0073] S44. Based on the correction evaluation results and the compliance of the constraints, perform non-dominated sorting on all individuals in the initial population to obtain the hierarchically sorted population sequence. Based on the correction and evaluation results output by S43, the system simultaneously verifies whether all constraints, such as geometry, structure, stability, draft, and navigation conditions, are met for each individual in the population. According to the multi-objective non-dominated relationship, all individuals in the initial population are divided into superior and inferior levels, and all individuals are assigned to different levels of Pareto non-dominated fronts. Finally, the system outputs an ordered population sequence with complete hierarchical sorting, providing a hierarchical basis for subsequent individual selection.

[0074] S45. Based on the objective dimension of the multi-objective optimization function, generate a uniformly distributed set of reference vectors in the objective space to obtain the reference vector dataset. Based on the performance index dimensions contained in the multi-objective optimization function constructed by S3, a set of uniformly distributed reference vectors is generated in the multi-objective performance space to obtain the reference vector dataset. This reference vector dataset serves as the benchmark for the NSGA-III algorithm, used to measure the density of individual distributions in the objective space, ensuring that the subsequent optimization solution set can uniformly cover each performance trade-off interval.

[0075] S46. Based on the population sequence and reference vector dataset, calculate the spatial distance between each individual and the reference vector in the same non-dominated level, and perform secondary screening on individuals in the same level based on the spatial distance to obtain a set of high-quality individuals. Based on the population sequence after hierarchical sorting in S44 and the reference vector dataset generated in S45, for all individuals in the same non-dominated hierarchy, the spatial distance between the individual's objective function point and each reference vector is calculated one by one. Individuals that are overly clustered or redundant in the target space are eliminated according to the spatial distance, while samples with balanced distribution and better overall performance are retained. Finally, a simplified set of high-quality individuals is obtained to maintain the diversity of Pareto solution sets.

[0076] S47 iteratively sorts the set of high-quality individuals based on genetic operators, judges the convergence status through convergence criteria, and outputs the optimal design parameters.

[0077] A complete genetic iteration operation is performed on the set of high-quality individuals output by S46. Tournament selection, simulated binary crossover, and polynomial mutation operators are used sequentially to complete parent selection, code recombination, and parameter perturbation. After merging the parent and offspring populations, the non-dominated sorting and individual selection process is repeated. After each iteration, the convergence criteria of the Pareto front change rate and the maximum number of iterations are verified. If the convergence condition is met, the optimal integrated ship-propeller-rudder design parameters are directly output. If the convergence condition is not met, steps S42 to S47 are repeated for continuous optimization. That is, the overall change rate of the Pareto front for consecutive iterations is calculated and compared with a preset convergence threshold. If the change rate is lower than the preset convergence threshold, the optimization process is considered converged. The current optimal solution set of the Pareto front is extracted, and the optimal design parameters for the hull, propeller, and rudder that take into account flow field coupling performance and multiple operating condition constraints are output. If the rate of change is higher than the preset convergence threshold, the current number of iterations is compared with the maximum number of iterations. If the maximum number of iterations has not been reached, return to S42 to start a new round of iterations. If the maximum number of iterations has been reached, extract the current Pareto front optimal solution set and output the optimal design parameters.

[0078] S5 uses a dynamic feedback mechanism to correct design parameters in real time during the optimization iteration process. The design parameters include model parameters, optimization parameters, and operating condition constraint parameters.

[0079] During the optimization process, the prediction accuracy of the surrogate model and changes in operating constraints are monitored in real time, and the optimization process is dynamically adjusted. Specifically, when the surrogate model's prediction error exceeds the limit, local samples are sparse, or there are insufficient new operating condition samples, adaptive sampling is triggered. Representative sample points are selected, and high-fidelity CFD calculations are performed to obtain the corresponding flow field performance indicators. Simultaneously, the high-fidelity CFD calculation results are added to the sample database for incremental training or retraining of the surrogate model, thereby improving its prediction accuracy and generalization ability. When using the RBF surrogate model, dynamic correction of the surrogate model is achieved by supplementing samples, adjusting kernel function parameters, and updating weight coefficients and error correction terms.

[0080] During the optimization iteration process, if a certain objective function continuously deteriorates, population diversity is insufficient, the Pareto front stagnates, or a significant local convergence trend occurs, the optimization parameters are dynamically adjusted. These parameters include population size, crossover probability, mutation probability, reference point distribution density, elite individual retention ratio, and high-fidelity correction sample ratio. In the preferred implementation using the NSGA-III algorithm, adjusting the reference point distribution, crossover probability, and mutation probability can enhance the search capability for high-potential regions and prevent the population from getting trapped in local optima.

[0081] Automatically update constraint functions when switching between different operating scenarios And adjust the target weights in a timely manner to achieve a balance between performance optimization and adaptability to operating conditions.

[0082] Please see Figure 3 and Figure 4 The diagram shown is a schematic of a ship propeller-rudder integrated design optimization system that incorporates flow field coupling effects, according to an embodiment of this application. The ship propeller-rudder integrated design optimization system includes: The ship parametric modeling module is used to uniformly model the hull parameters, propeller parameters, and rudder parameters, and obtain an integrated geometric model of the ship, propeller, and rudder through surface fitting. The ship parametric modeling module enables synchronous modification of ship-propeller-rudder geometry through a unified parameter interface, providing a flexible design space for global optimization and a high-quality geometric model for subsequent flow field coupling calculations, including hull modeling units, propeller modeling units, and rudder surface modeling units. The hull modeling unit uses B-spline curves and feature line constraints to fit hull parameters and generate hull surfaces. Hull parameters include length L, beam B, draft T, stern streamline control points, and waterline parameters. The propeller modeling unit is used to model propeller parameters to determine the profile and spatial position of each blade section, and to generate new propeller blade surfaces through surface fitting. Propeller parameters include the number of blades Z, pitch ratio P / D, and aspect ratio. and leaf tip thickness ; The rudder surface modeling unit is used to determine the rudder surface profile, rudder section position, and installation attitude through rudder surface parameters, and then generate a new rudder surface model through fitting.

[0083] After each unit is parametrically modeled, the ship hull parametric modeling module generates an integrated geometric model of the ship, propeller, and rudder based on the fitted hull surface, propeller blade, and rudder surface models, which is used for subsequent flow field coupling calculations.

[0084] The flow field coupling calculation module is used to perform flow field coupling calculations on the hull, propeller and rudder based on the geometric model, and obtain the performance index data corresponding to each working condition. The flow field coupling calculation module includes a CFD calculation unit, a working condition adaptation unit, and a data output unit; The CFD calculation unit is used to analyze the flow field distribution at the stern, propeller disk load and rudder surface force based on the geometric model, and to perform flow field coupling calculations on the hull, propeller and rudder surface to obtain the performance index data corresponding to each working condition. The operating condition adaptation unit is used to divide the operating conditions according to the actual operating characteristics of the ship and match the operating condition conditions with the CFD calculation unit. The data output unit is used to output the performance index data corresponding to each operating condition. The performance index data includes hull drag coefficient, propeller induced drag, rudder additional drag, propulsion efficiency, seakeeping performance evaluation index, maneuvering performance evaluation index, and local pressure pulsation.

[0085] The operating condition constraint management module is used to transform the ship's operating characteristics under different navigation conditions into design constraints and to map complex operating environments into design boundary conditions, thereby obtaining the constraint conditions for different operating conditions. The operating condition constraint management module can ensure that the optimization results meet the requirements of seaworthiness, structural safety and operational performance, so that the optimization scheme has adaptability and stability under various navigation conditions, including shallow water operating condition constraint unit, confined water area operating condition constraint unit and deep water wave operating condition constraint unit. Shallow water condition constraint element is used to construct constraint conditions for shallow water conditions in order to evaluate the impact of shallow water effects on hull resistance, propulsion efficiency and propeller front flow field distribution. The confined waters operating condition constraint unit is used to construct the constraint conditions for confined waters operating conditions in order to evaluate the ship's low-speed maneuverability, rudder effectiveness, and flow disturbances between the ship, propeller, and rudder. The deep-water wave condition constraint element is used to construct constraint conditions for deep-water wave conditions in order to evaluate ship propulsion performance, seakeeping performance, and local pressure pulsation.

[0086] The rapid optimization and optimization module is used to perform nonlinear regression prediction of performance indicators under different combinations of design variables by constructing a surrogate model based on performance index data and constraints, and to perform global optimization in the design variable space based on a multi-objective optimization algorithm to obtain the optimal design parameters. The rapid optimization and optimization module is used to efficiently obtain the optimal design scheme of the ship-propeller-rudder integration under global constraints. It reduces the calculation cost of CFD high-fidelity flow field by building a surrogate model and performs global optimization in the design variable space through a multi-objective optimization algorithm, so that the final scheme can maintain better propulsion efficiency, maneuverability and economy under different working conditions. It includes a surrogate model building unit, a global optimization unit and a high-precision correction unit. The proxy model building unit is used to build a proxy model to perform nonlinear regression prediction of performance indicators under different combinations of design variables, and obtain the predicted values ​​of performance indicators for each working condition. The global optimization unit is used to determine the optimization objective based on the actual operating characteristics of the ship under different operating conditions, construct a multi-objective optimization function in combination with the operating condition constraints, and perform global optimization in the design variable space based on the optimization algorithm to obtain the optimal design parameters. The high-precision correction unit is used to calculate the deviation between the performance index value obtained from the actual flow field coupling and the performance index prediction value predicted by the surrogate model, and correct the surrogate model based on the deviation.

[0087] The dynamic feedback module uses a dynamic feedback mechanism to correct design parameters in real time during the optimization iteration process. Design parameters include design variables, model parameters, optimization parameters, and operating condition constraint parameters.

[0088] The dynamic feedback module embodies the system's self-learning capability, which can improve convergence speed and result stability. It includes an efficiency monitoring unit, a parameter adjustment unit, and a feedback interface unit. The efficiency monitoring unit is used to monitor performance indicators in real time, adaptively optimize the surrogate model based on the deviation between the actual monitored values ​​and the predicted values, and dynamically switch operating condition constraints to obtain monitoring results. The parameter adjustment unit is used to dynamically adjust design variables, model parameters, optimization parameters, and operating condition constraint parameters to obtain adjustment results and avoid the optimization process from getting trapped in local optima. The feedback interface unit is used to feed back the monitoring results and adjustment results to the flow field coupling calculation module, the operating condition constraint management module, and the rapid optimization and optimization module in real time.

[0089] The solution of this application will be described in detail below with reference to specific embodiments.

[0090] This embodiment uses a single-propeller, single-rudder transport vessel as the optimization object, and performs integrated design optimization of its stern hull lines, propeller parameters, and rudder surface parameters. The transport vessel has a length L of 82m, a beam B of 13.5m, a design draft T of 3.0m, and a propeller blade count Z of 4. Considering the different restricted operating conditions that the vessel may experience during actual operation, such as shallow water navigation, low-speed maneuvering in confined waters, and deep-water wave navigation, this embodiment sets the following three representative navigation conditions: The first operating condition is shallow water navigation, with a ship speed of 10 knots and a water depth of 5 meters. It is mainly used to evaluate the impact of shallow water effect on hull resistance, propulsion efficiency and propeller front flow field distribution. The second operating condition is the navigation condition in restricted waters, with a ship speed of 6 knots, an effective channel width of 5 times the ship's beam B, and a channel depth of 6 meters. It is mainly used to evaluate the ship's low-speed maneuverability, rudder effectiveness, and flow interference between the ship, propeller, and rudder. The third operating condition is the deep-water wave navigation condition, with a ship speed of 12 knots, a water depth of 30 meters, a wave height of 1.5 meters, and a wave period of 6 seconds. It is mainly used to evaluate the ship's propulsion performance, seakeeping performance, and local pressure pulsation.

[0091] While keeping the ship's main dimensions (length L, beam B, draft T, and number of propeller blades Z) constant, select the stern streamline control point, waterline parameters, pitch ratio P / D, and aspect ratio. Leaf tip thickness Rudder area coefficient Longitudinal mounting position of the rudder and rudder tilt angle As design variables, the resulting design variable space is represented as follows: ;in, These are the streamline control point parameters for the stern, with values ​​ranging from 0.97 to 1.03. This is a waterline profile parameter, with a value range of 0.97 to 1.03; The pitch ratio P / D ranges from 0.88 to 1.04. for aspect ratio The value ranges from 0.55 to 0.70; Thickness at the leaf tip The value ranges from 10.8 to 13.2 mm; Rudder area coefficient The value ranges from 0.018 to 0.022; The longitudinal mounting position of the rudder The value ranges from 0.972 to 0.986; Rudder angle The value range is -1.5° to 1.5°.

[0092] Within the constructed design variable space, the Latin hypercube sampling method is used to generate initial training samples. In this embodiment Set to 100. For each set of design variables, reconstruct the corresponding hull, propeller, and rudder geometry models, and import the hull, propeller, and rudder into a unified computational domain according to their actual installation positions.

[0093] Within a unified computational domain, the regions containing the hull and rudder are designated as static flow fields, while the region containing the propeller is designated as a rotating flow field. Flow field data, including velocity, pressure, and turbulence parameters, are transferred through the interface between these regions. During the coupled flow field calculation, the non-uniform wake generated at the stern of the hull serves as the actual inflow condition for the propeller, affecting the velocity distribution, thrust, torque, and induced flow characteristics at the propeller disk. The rotating induced flow and accelerating flow generated by the propeller further act on the rudder, altering the velocity and pressure distribution around the rudder and the load on the rudder surface. Simultaneously, the flow field changes caused by propeller rotation and rudder force react on the stern of the hull, further altering the pressure distribution and wake characteristics at the stern. During the calculation, the flow fields around the hull, the rotating propeller flow field, and the flow fields around the rudder are iteratively coupled and solved. In each iteration, the flow field data is updated and exchanged between the static flow field region and the rotating flow field region until the hull resistance, propeller thrust and torque and rudder load reach a stable state, thereby obtaining the ship-propeller-rudder coupled flow field under the interaction of the hull wake, propeller induced flow and rudder force.

[0094] The performance indicators corresponding to each operating condition are obtained and represented as follows: ; In the formula, m represents the navigation condition number, m=1, 2, 3; Indicates the hull drag coefficient. Indicates the induced drag of the propeller. Indicates the additional drag of the rudder. Indicates the efficiency of advancement. Indicates the evaluation index of wave resistance performance. Indicates the performance evaluation index of handling. This indicates localized pressure pulsation.

[0095] Using performance index data corresponding to different combinations of design variables obtained from coupled calculations as sample data, RBF surrogate models were established between the design variables and various performance indices under three operating conditions. Nonlinear regression approximations were performed on hull resistance, propulsion efficiency, rudder effectiveness, seakeeping performance, maneuverability, and flow field coupling indices. The K-fold cross-validation method was used to test the prediction error of the RBF surrogate models. When the prediction error When the error is less than 5%, the RBF surrogate model will be used for subsequent optimization; when the prediction error... When the error exceeds 5%, CFD calculation samples are added to areas with large prediction errors or sparse sample distribution, and the RBF surrogate model is updated.

[0096] Based on the operational requirements under three working conditions, a multi-working-condition optimization objective is established. The first optimization objective is to minimize the combined ship-propeller-rudder drag. ; The second optimization objective is to maximize the integrated propulsion efficiency under multiple operating conditions: ; The third optimization objective is to maximize seakeeping performance under deep-water wave navigation conditions: ; The fourth optimization objective is to maximize maneuverability under navigation conditions in confined waters: ; The fifth optimization objective is to minimize local pressure fluctuations under deep-water wave navigation conditions: ; This leads to a multi-objective optimization problem, expressed as: ; Among them, the evaluation proportions of shallow water navigation, restricted water navigation, and deep water wave navigation in the comprehensive resistance and comprehensive propulsion efficiency targets are 0.35, 0.25, and 0.40, respectively.

[0097] Set wake uniformity constraints during optimization. Stability constraints Structural constraints and draft and operating condition constraints Among them, under shallow water navigation conditions, the non-uniformity of the flow velocity distribution in front of the propeller does not exceed a preset threshold; under navigation conditions in confined waters, the maneuverability evaluation index... Not lower than the preset threshold; wave resistance performance evaluation indicators under deep-water wave navigation conditions. Local pressure pulsation not lower than a preset threshold The maximum stress shall not exceed the preset upper limit; under all operating conditions, the maximum stress of the blades and control surfaces shall not exceed the allowable stress of the corresponding materials.

[0098] Then, the NSGA-III algorithm, supported by the RBF surrogate model, is used for multi-objective optimization, with population size... Set to 100, maximum number of iterations. Set to 120. For each individual in the population, call the RBF proxy model corresponding to the three working conditions respectively to predict its performance index under different working conditions, and calculate... During the optimization process, every 10 iterations, design schemes near the Pareto front, with large prediction errors in the RBF surrogate model, or close to the constraint boundaries are selected for high-precision CFD correction. When the relative error between the CFD calculation results and the RBF surrogate model prediction results exceeds 5%, the corresponding results are added to the sample database and the RBF surrogate model is updated. When a design scheme performs well under one operating condition but violates constraints or exhibits significant performance degradation under other operating conditions, the optimization process is shifted towards a region with relatively balanced performance across multiple operating conditions by adjusting the target weights, reference point distribution, and optimization parameters. After optimization, a Pareto optimal solution set that simultaneously satisfies the constraints of all three operating conditions is obtained, and a compromise scheme with better overall performance is selected from this set.

[0099] In the compromise solution selected in this embodiment, The corresponding parameters are: stern streamline control point parameter 0.98, waterline profile parameter 1.02, pitch ratio P / D 0.96, and aspect ratio 0.98. The thickness at the leaf tip is 0.63. It is 11.6mm, and the rudder area coefficient is... The longitudinal mounting position of the rudder is 0.0208. The rudder tilt angle is 0.979. for High-precision CFD calculations verified that, compared to the initial design, this compromise solution reduces the weighted total drag of the ship, propeller, and rudder by approximately 4.3% in shallow water navigation conditions, while also improving propulsion efficiency. Improved by approximately 3.8%; Maneuverability evaluation indicators under restricted water navigation conditions Increased by approximately 6.5%, additional rudder drag. Reduced by approximately 3.2%; propulsion efficiency under deep-water wave navigation conditions. Improved by approximately 4.6%, wave resistance performance evaluation index Increased by approximately 5.1%, local pressure pulsation It decreased by approximately 12.8%.

[0100] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0101] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0102] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0103] Please see Figure 5 , Figure 5 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 501 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 502 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 502 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 502 and is called and executed by the processor 501 using the methods described in the embodiments of this application. The input / output interface 503 is used to implement information input and output; The communication interface 504 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 505 transmits information between various components of the device (e.g., processor 501, memory 502, input / output interface 503, and communication interface 504); The processor 501, memory 502, input / output interface 503, and communication interface 504 are connected to each other within the device via bus 505.

[0104] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0105] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0106] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0107] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0108] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0109] This application provides a method and system for integrated design optimization of ship, propeller, and rudder by introducing flow field coupling effects. This scheme models the hull, propeller, and rudder surfaces with unified parameters and generates an integrated geometric model through surface fitting. By unifying the parameters of the three types of components into the same modeling system, it eliminates geometric gaps and flow field interface mismatches caused by separate component modeling, fully restoring the spatial assembly position, external surface shape, and the true geometric boundaries of mutual interference between the ship, propeller, and rudder. Based on this geometric model, flow field coupling calculations of the ship, propeller, and rudder can accurately capture the mutual flow field coupling effects between the hull wake, propeller wake vortex, and rudder surface disturbance, obtaining high-precision sample performance data. By constructing a surrogate model, nonlinear regression prediction of design variables and multi-condition performance indicators is achieved. Low-cost, fast regression prediction replaces full-scale, time-consuming CFD simulation, significantly reducing the computational cost of performance calculations under multiple variables and conditions, and rapidly outputting performance indicators corresponding to any combination of design variables in batches.

[0110] By combining the actual navigation conditions of the ship, multiple optimization objectives are determined, and a multi-objective optimization function is constructed by integrating various operational boundary constraints. Abandoning the limitations of single-performance index optimization, and aligning with the actual operational needs of the ship under all navigation conditions, multiple optimization objectives are simultaneously determined. At the same time, ship geometric constraints, structural strength, stability, draft, and operational limitations under different speeds / load conditions are transformed into standardized constraints. This constructs a multi-objective optimization function that considers multiple performance requirements and fits the actual ship operational boundaries, ensuring that the subsequent optimization process is always carried out within a feasible design space that meets the actual usage requirements of the ship, avoiding optimization results that deviate from engineering reality. Based on the multi-objective optimization function, surrogate model, and predicted performance indicators, a multi-objective optimization algorithm is used to globally optimize in the design variable space, outputting optimal design parameters. By traversing a massive number of design parameter combinations, the Pareto optimal solution set that considers multiple conflicting performance objectives is automatically selected, escaping local optima and extensively exploring ship-propeller-rudder matching schemes with superior overall performance within the design variable space, achieving integrated initial selection of schemes under multiple variables, constraints, and objectives.

[0111] A dynamic feedback closed-loop mechanism is established to correct various parameters in real time during the iterative optimization process, continuously compensating for prediction deviations caused by flow field coupling. By correcting relevant parameters of the geometric model, the prediction errors between the surrogate model and the actual CFD calculation are compensated. By adjusting optimization parameters such as the population size and number of iterations of the optimization algorithm, the global search balance is improved. By dynamically fine-tuning the constraint thresholds of various navigation conditions, the performance boundary changes caused by flow field coupling are adapted. The prediction deviation of the surrogate model, the search defects of the optimization algorithm, and the mismatch of the operating condition constraints are continuously weakened. The optimal parameters are continuously refined through iteration, and finally, a ship-propeller-rudder integrated design scheme with higher flow field coupling effect matching, stronger adaptability to all operating conditions, and better engineering feasibility is obtained.

[0112] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0113] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0114] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0115] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0116] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for optimizing the integrated design of ship propeller and rudder by incorporating flow field coupling effects, characterized in that, The method includes: The hull parameters, propeller parameters, and rudder surface parameters are modeled in a unified manner, and a geometric model integrating the ship, propeller, and rudder is obtained through surface fitting. Based on the geometric model, flow field coupling calculations are performed on the hull, propeller and rudder, and nonlinear regression predictions are performed on the performance indicators under different combinations of design variables by constructing a surrogate model to obtain the performance indicators corresponding to each working condition. The optimization objective is determined based on the actual operating characteristics of the ship under different operating conditions, and a multi-objective optimization function is constructed in combination with the operating condition constraints. Based on the multi-objective optimization function, the surrogate model, and the performance index, the optimal design parameters are obtained by performing global optimization in the design variable space based on the optimization algorithm. The design parameters are corrected in real time during the optimization iteration process based on the dynamic feedback mechanism. The design parameters include design variables, model parameters, optimization parameters and operating condition constraint parameters.

2. The ship propeller-rudder integrated design optimization method according to claim 1, characterized in that, The design variables include stern streamline control point parameters, waterline parameters, pitch ratio, aspect ratio, blade tip thickness, rudder area coefficient, rudder longitudinal installation position, and rudder inclination angle; the performance indicators include hull drag coefficient, propeller induced drag, rudder additional drag, propulsion efficiency, seakeeping performance evaluation indicators, maneuverability evaluation indicators, and local pressure pulsation.

3. The ship propeller-rudder integrated design optimization method according to claim 1, characterized in that, Based on the aforementioned geometric model, flow field coupling calculations are performed on the hull, propeller, and rudder surfaces. Furthermore, a surrogate model is constructed to perform nonlinear regression predictions of performance indicators under different combinations of design variables, yielding performance indicators for each operating condition, including: The design variables are determined based on the parameters and a design variable space is constructed. Training samples are then generated within the design variable space. Based on the training samples, flow field coupling calculations were performed on the hull, propeller and rudder to obtain the corresponding performance indicators under different operating conditions. Establish a proxy model between the design variables and the performance indicators, and train the proxy model based on the training samples to obtain the trained proxy model; Based on the trained surrogate model, nonlinear regression prediction is performed on the performance indicators to obtain the predicted values ​​of the performance indicators for each working condition.

4. The ship propeller-rudder integrated design optimization method according to claim 3, characterized in that, Based on the training samples, flow field coupling calculations were performed on the hull, propeller, and rudder to obtain performance indicators under different operating conditions, including: A unified computational domain is established based on the actual installation positions and relative spatial relationships of the hull, propeller, and rudder. Within the unified computational domain, the inner-layer physical flow field coupling calculation is performed on the flow field around the hull, the rotating flow field of the propeller, and the flow field around the rudder surface to obtain the coupling calculation results. The flow field characteristics and performance indicators are extracted from the coupling calculation results, and the flow field characteristics and performance indicators are compared with the preset target values ​​or constraint thresholds under the corresponding navigation conditions to obtain the matching deviation. The design variables are iteratively adjusted based on the matching deviation to regenerate the integrated geometric model of the ship, propeller, and rudder. The integrated geometric model is recalculated for flow field coupling until the performance indicators corresponding to each working condition under steady-state conditions are obtained.

5. The ship propeller-rudder integrated design optimization method according to claim 4, characterized in that, The flow field data transfer process within the unified computing domain includes: Determine the static flow field region, the rotating flow field region, and the region interface within the unified computational domain; Computational grids are generated for the static flow field region and the rotating flow field region respectively, and donor grid cells and acceptor grid cells are determined according to the spatial positional relationship between the grid cells and the interface of the region. The interpolation weights are determined based on the donor grid cells and the recipient grid cells; Based on the propeller speed and the position of the grid points at the region interface relative to the propeller axis, the velocity coordinates of the static flow field region and the rotating flow field region are transformed. Based on the velocity coordinate transformation and the interpolation weight, the flow field parameters are transferred within the unified computational domain. The flow field data includes velocity, pressure, turbulence parameters, and flux.

6. The ship propeller-rudder integrated design optimization method according to claim 4, characterized in that, The process of determining steady-state conditions during flow field coupling calculations includes: By comprehensively using the residuals of the governing equations, the mass conservation error of the regional interface, and hydrodynamic monitoring data, the stability state of the inner physical flow field coupling is determined, and the first determination result is obtained. After the coupling calculation of the inner physical flow field reaches a stable state, the propulsion balance relationship between the propeller thrust and the total resistance generated by the hull, rudder surface and other appendages is judged to obtain the second judgment result. When both the first and second determination results are stable, the determination reaches the steady-state condition. When the steady-state condition is reached, the dynamic matching state of the geometric model and the flow field coupling calculation results is determined, and the feedback adjustment of the current design scheme is stopped when the dynamic matching state is reached.

7. The ship propeller-rudder integrated design optimization method according to claim 1, characterized in that, The operating conditions include shallow water navigation, restricted water navigation, and deep water wave navigation; the constraints include wake uniformity constraints, stability constraints, structural constraints, and draft and operating condition constraints.

8. The ship propeller-rudder integrated design optimization method according to claim 1, characterized in that, The optimization algorithm is the NSGA-III algorithm. Based on the multi-objective optimization function and the surrogate model, the NSGA-III algorithm performs global optimization in the design variable space to obtain the preferred design parameters, including: The population size is set according to the ship design variables and their range, and an initial population is randomly generated. Based on the initial population, the proxy model is invoked to perform nonlinear regression prediction on the performance indicators corresponding to each combination of design variables to obtain fitness evaluation results. Key individuals are selected from the initial population and ship propeller-rudder flow field coupling calculations are performed. The fitness evaluation results are corrected based on the actual calculated performance index values ​​to obtain the corrected evaluation results. Based on the correction and evaluation results, and considering the compliance of the constraints, all individuals in the initial population are sorted in a non-dominated manner to obtain the hierarchical sorted population sequence. Based on the objective dimension of the multi-objective optimization function, a uniformly distributed set of reference vectors is generated in the objective space to obtain a reference vector dataset. Based on the population sequence and the reference vector dataset, the spatial distance between each individual and the reference vector in the same non-dominated level is calculated. Based on the spatial distance, individuals in the same level are screened again to obtain a set of high-quality individuals. The set of high-quality individuals is iteratively sorted based on genetic operators, and the convergence status is determined by convergence criteria to output the optimal design parameters.

9. A ship propeller-rudder integrated design optimization system incorporating flow field coupling effects, applied to the ship propeller-rudder integrated design optimization method according to any one of claims 1 to 8, characterized in that, The integrated ship propeller and rudder design optimization system includes: The ship parametric modeling module is used to uniformly model the hull parameters, propeller parameters, and rudder parameters, and obtain an integrated geometric model of the ship, propeller, and rudder through surface fitting. The flow field coupling calculation module is used to perform flow field coupling calculations on the hull, propeller and rudder based on the geometric model, and obtain the performance index data corresponding to each working condition. The operating condition constraint management module is used to transform the ship's operating characteristics under different navigation conditions into design constraints and to map complex operating environments into design boundary conditions, thereby obtaining the constraint conditions for different operating conditions. The rapid optimization and optimization module is used to perform nonlinear regression prediction of performance indicators under different combinations of design variables by constructing a surrogate model based on performance index data and constraints, and to perform global optimization in the design variable space based on a multi-objective optimization algorithm to obtain the optimal design parameters. The dynamic feedback module, based on the dynamic feedback mechanism, corrects the design parameters in real time during the optimization iteration process. The design parameters include design variables, model parameters, optimization parameters, and operating condition constraint parameters.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 8.

Images