A design method for hydrodynamic performance optimization of semi-submersible wind power platform

By optimizing the key dimensions of the semi-submersible wind power generation platform through Sobol global sensitivity analysis and NSGA-III algorithm, the problems of long design cycle and high cost in the existing technology are solved, and efficient and robust hydrodynamic performance optimization is achieved, thereby improving the stability and efficiency of the platform.

CN123069982APending Publication Date: 2026-11-10DONGFANG ELECTRIC(FUJIAN)INNOVATION INST CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611381195.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-07
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing design methods cannot systematically consider the complex interactions between different geometric dimensions of semi-submersible wind power platforms, resulting in long design cycles, high costs, and iterative processes, and failing to effectively optimize hydrodynamic performance.

Method used

Sobol global sensitivity analysis was used to screen key parameters, and the three parameters were adjusted in conjunction with the NSGA-III algorithm. Frequency domain potential flow theory was used to optimize the hydrodynamic performance of the platform. Stability was ensured by narrowing the design range and hydrostatic verification. Key dimensions were optimized to minimize the response amplitude operator.

Benefits of technology

Significantly shorten the design cycle, reduce computational risks, improve the hydrodynamic performance and stability of the platform, and optimize the design scheme to enhance the actual operating efficiency and safety of the semi-submersible wind power platform.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present application relates to a kind of design methods for the hydrodynamic performance optimization of semi-submersible wind power platform, it starts to determine a benchmark model, and selects key main size as design variable.Subsequently, using Sobol global sensitivity analysis method, to identify the design variable that is most significant to hydrodynamic performance.In through hydrodynamic school to narrow the design range after school, final optimization is carried out.This optimization step utilizes the six degrees of freedom motion response amplitude operator (RAO) of platform calculated by frequency domain potential flow theory, and by iteration and linkage adjustment to three key size parameters, aims at minimizing platform motion response, especially pitch and heave motion, so as to improve the overall stability and operating efficiency of entire wind turbine.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of semi-submersible wind power generation platform modeling technology, and specifically to a design method for optimizing the hydrodynamic performance of semi-submersible wind power generation platforms. Background Technology

[0002] Offshore wind energy is a rapidly developing branch of renewable energy. For water depths exceeding 50-60 meters, fixed foundations are no longer economically feasible, making floating platforms a necessary option. Among various floating platform concepts (such as Spar, tension leg platforms (TLP), and semi-submersible platforms), semi-submersible platforms are favored for their superior stability, relatively simple installation process, and applicability across a wide range of water depths.

[0003] The hydrodynamic performance of a semi-submersible platform is crucial to the safety, efficiency, and feasibility of the entire wind turbine system. Excessive platform motion, especially pitch (rotation about the horizontal axis) and heave (vertical motion), can reduce power generation efficiency and impose fatigue loads on the turbine structure, potentially leading to structural failure in extreme cases.

[0004] Current design methods typically rely on trial and error or single-variable optimization, such as the "Main Dimension Design Method for Offshore Wind Power Floating Platform" disclosed in patent number CN 117763649 A, or the "Parameter Design Method for Floating Wind Turbine Semi-Submersible Platform" disclosed in patent number CN117744515A. These methods often fail to systematically consider the complex interactions between different geometric dimensions of the platform. In the early stages of design, the sensitivity of hydrodynamic response to dimensional changes is often not fully understood, leading to costly and iterative design iteration cycles.

[0005] Therefore, there is an urgent need for a more systematic, efficient, and robust design methodology that can effectively identify and optimize the most critical geometric dimensions to improve the hydrodynamic performance of semi-submersible wind power platforms. Such a methodology should shorten the design cycle, reduce computational risks, and ultimately result in a platform with superior stability and performance under actual operating conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a design method for optimizing the hydrodynamic performance of semi-submersible wind power generation platforms to solve the aforementioned problems.

[0007] This invention provides the following technical solution: A design method for optimizing the hydrodynamic performance of a semi-submersible wind power platform includes the following steps: Select the basic layout; select the basic layout of the semi-submersible wind power generation platform. Initial parameters and hydrodynamic model: Determine the initial principal dimension parameters of the semi-submersible wind power platform and construct the hydrodynamic model; Filter key design variables, select key parameters from the main dimension parameters for subsequent optimization, and determine the design range of each key parameter; Sobol global sensitivity analysis uses the Sobol global sensitivity analysis method to perform sensitivity analysis on selected key parameters, quantifies the influence of each key parameter, and selects the N key parameters with the highest sensitivity. Narrow down the design scope and filter out designs that do not meet the basic hydrostatic requirements; Using frequency domain potential flow theory, the objective function is to minimize the response amplitude operator of six degrees of freedom. The minimization of the response amplitude operator includes maximizing the heave period, minimizing the heave motion response, and minimizing the pitch motion response. The NSGA-III algorithm is used to adjust N key parameters in a coordinated manner and iterate repeatedly until a design configuration that can minimize the response amplitude operator to the target level is found, thereby obtaining an optimized design scheme.

[0008] Preferably, the parameters are based on environmental data of the target wind field and the parameters of the selected wind turbine, wherein the environmental data of the target wind field includes wave height, wave period and wind speed, and the parameters of the wind turbine include weight, hub height and thrust.

[0009] Preferably, the hydrodynamic model is constructed using industry-standard frequency domain potential flow software.

[0010] Preferably, the key parameters in the main dimension parameters include column diameter Dc, buoy height Hp, buoy width Bp, platform draft T, and column spacing Ls.

[0011] Preferably, the N key parameters with the highest sensitivity are selected, including column diameter Dc, platform draft T, and pontoon height Hp.

[0012] Preferably, the NSGA-III algorithm is used to adjust N key parameters in a coordinated manner, including: Parameter initialization: First, determine the initial value range of the column diameter Dc, platform draft T, and buoy height Hp. These parameters serve as optimization variables for the NSGA-III algorithm. Population initialization: Randomly generate an initial population within the parameter value range, with each individual representing a set of parameter combinations; Fitness assessment: The fitness of each individual is assessed according to the objective function. The heave period, heave motion response and pitch motion response under each parameter combination are calculated to determine the fitness value of the individual. Selection operation: A selection mechanism based on non-dominated sorting is adopted to divide the population into different levels and prioritize the selection of non-dominated individuals into the next generation; Crossover operation: Perform crossover operation on selected individuals to generate new individuals, and explore new solution spaces by exchanging some gene information; Mutation operation: Perform mutation operation on newly generated individuals to change some of the individual's gene values ​​with a certain probability; Environmental selection: The new individuals resulting from crossover and mutation are merged with the parent population, and non-dominated sorting and crowding calculations are performed again to select individuals with high fitness to form the next generation population; Iterative optimization: Repeat the above selection, crossover, mutation, and environment selection process until the preset number of iterations is reached or the convergence condition is met.

[0013] Preferably, the crossover operation uses simulated binary crossover; the mutation operation uses methods such as polynomial mutation to increase the diversity of the population and prevent the algorithm from getting trapped in local optima.

[0014] Preferably, the design range of the key parameters is defined based on the actual lower and lower limits of manufacturing constraints, structural requirements, and engineering experience.

[0015] Preferably, the narrowing of the design scope involves performing hydrostatic calculations on a series of promising designs identified from the Sobol global sensitivity analysis, and filtering out designs that do not meet the basic hydrostatic requirements.

[0016] Preferably, the narrowing of the design scope also includes checking stability, freeboard, draft, and displacement.

[0017] The present invention has the following beneficial technical effects: This invention uses Sobol global sensitivity analysis to quantify the impact of each design variable and its interactions on the hydrodynamic response, thereby focusing on the parameters with the greatest impact. Subsequently, hydrostatic kernel analysis is performed to ensure that potential design schemes meet basic stability requirements, thereby filtering out infeasible schemes and narrowing down the design space for the next stage. Finally, the six-degree-of-freedom response amplitude operator of the frequency domain potential flow theory calculation platform is used.

[0018] Optimization is achieved through an innovative "three-parameter linkage adjustment," which involves collaboratively modifying the three most sensitive dimensional parameters to minimize the peak value of the platform's response amplitude operator at key wave frequencies, thereby significantly improving the platform's stability.

[0019] This invention can effectively identify key design parameters and efficiently and robustly optimize the platform shape to achieve superior hydrodynamic performance. Detailed Implementation

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

[0021] The Ratio of Amplitudes Output (RAO) is a widely used operator in signal processing. Its main function is to calculate the ratio of the amplitudes of two signals to measure the difference in signal intensity.

[0022] Example: S1, Determine the baseline model and select design variables. ① Select a basic layout: Based on the environmental data of the target wind field (such as wave height, wave period, and wind speed) and the parameters of the selected wind turbine (such as weight, hub height, and thrust), select a basic structural layout. This selection refers to the semi-submersible platform design that has been validated in industry and academia, which is usually composed of three or four vertical columns connected by underwater horizontal pontoons and / or support structures.

[0023] ② Initial Parameters and Hydrodynamic Model: The initial principal dimensions of the semi-submersible platform were determined. Subsequently, a hydrodynamic model was constructed using industry-standard frequency domain potential flow software.

[0024] ③ Screening Key Design Variables: Select several key parameters from the list of master dimension parameters for subsequent optimization. These key parameters are considered to significantly affect the platform's hydrostatic and hydrodynamic characteristics. In a preferred embodiment, these variables include, but are not limited to: column diameter Dc, pontoon height Hp, pontoon width Bp, platform draft T, and column spacing Ls; S2, Design Variable Initialization For each key parameter (design variable) selected in step S1, a design range or design space is defined. This design range or design space is defined based on the actual lower and lower limits of manufacturing constraints, structural requirements, and engineering experience. This ensures that subsequent optimization processes are explored only within feasible solutions.

[0025] S3, Perform local sensitivity analysis (optional) As a preliminary step, a local sensitivity analysis can be performed. In this analysis, only one design variable is varied at a time, and its effect on the hydrodynamic response (e.g., pitch RAO) is observed. This step provides a preliminary understanding of the influence of each parameter, but it cannot capture the interaction effects between parameters.

[0026] S4, Sobol Global Sensitivity Analysis Unlike local analyses, Sobol Global Sensitivity Analysis is a global sensitivity analysis technique based on variance decomposition. It assesses the impact of the independent contribution (first-order exponent) of each design variable and the contribution of interactions between variables (higher-order exponents) on the output variance.

[0027] Execution: Thousands of design configurations are generated within the scope defined by S2 using a sampling technique (such as Sobol sequence sampling). The hydrodynamic response is calculated for each configuration.

[0028] Analysis: Calculate the Sobol index. This quantifies the influence of each principal dimension parameter. For example, the analysis might reveal that the column diameter Dc and platform draft T are the main drivers of heave motion, while the column diameter Dc and pontoon height Hp strongly influence pitch motion. This step allows design variables to be ranked from most important to least important.

[0029] S5, Conduct hydrostatic calculation and verification analysis to narrow down the design scope. Before performing computationally intensive detailed hydrodynamic optimization, designs that do not meet the basic hydrostatic requirements are filtered out.

[0030] Calculations: Perform hydrostatic calculations on a range of promising designs identified from S4. This includes verification: Stability: The transverse metacentric height (GMT) must be within a safe range to ensure positive initial stability. The transverse metacentric height (GMT) measures the ship's ability to resist tilting (heeling).

[0031] Freeboard: There must be sufficient freeboard height to prevent waves from surging up.

[0032] Draft and Displacement: The displacement must be sufficient to support the total weight of the turbine-platform system.

[0033] Narrowing the scope: Designs that fail this check will be removed from the design space. Based on the results, the scope of design variables can be adjusted and narrowed, allowing subsequent optimization efforts to focus on more promising and stable solution regions.

[0034] S6, Optimization based on frequency domain potential flow theory Calculating the six-DOF RAO response: Using frequency domain potential flow theory, the response amplitude operator (RAO) for six degrees of freedom (heave, roll, pitch, yaw, sway, and sway) is calculated for the candidate design. RAO represents the motion response amplitude of a semi-submersible platform to waves of unit amplitude within a certain frequency range. The main objective is to minimize the RAO, particularly reducing the pitch and heave responses within the frequency range where wave energy is concentrated.

[0035] Optimization is achieved through a three-parameter coordinated adjustment: instead of adjusting one variable at a time, this method coordinates the adjustments of the three most sensitive key parameters identified in S4. For example, if the column diameter Dc, platform draft T, and buoy height Hp are determined to be the most influential, they are adjusted synergistically. This adjustment can be based on a predefined function or relationship to systematically manipulate the platform's natural periods (e.g., shifting the natural periods of pitch and heave away from the range of high-energy wave periods) and the counteracting effects of wave loads.

[0036] Specific optimization methods: The specific method for optimization through three-parameter linkage adjustment is as follows: The NSGA-III algorithm is used to link and adjust the three most sensitive parameters identified in S4 (column diameter Dc, platform draft T, and buoy height Hp). The objective functions include: a larger heave period is better, a smaller heave response is better, and a smaller pitch response is better. The specific optimization process is as follows: ① Parameter initialization: First, determine the initial value range of column diameter Dc, platform draft T, and buoy height Hp. These parameters serve as optimization variables for the NSGA-III algorithm.

[0037] ② Population initialization: Randomly generate an initial population within the parameter value range, with each individual representing a set of parameter combinations.

[0038] ③ Fitness Assessment: Fitness assessment is performed on each individual based on the objective function. The heave period, heave response, and pitch response are calculated for each parameter combination to determine the individual's fitness value.

[0039] ④ Selection operation: A selection mechanism based on non-dominated sorting is adopted, which divides the population into different levels and prioritizes non-dominated individuals to enter the next generation.

[0040] ⑤ Crossover operation: Perform a crossover operation on the selected individuals to generate new individuals. Crossover operations can employ methods such as simulated binary crossover (SBX), exploring new solution spaces by exchanging partial gene information.

[0041] ⑥ Mutation operation: Mutation operation is performed on newly generated individuals to change some gene values ​​of the individual with a certain probability. Mutation operation can use methods such as polynomial mutation to increase the diversity of the population and prevent the algorithm from getting trapped in local optima.

[0042] ⑦ Environmental selection: The new individuals after crossover and mutation are merged with the parent population, and non-dominated sorting and crowding calculation are performed again. Individuals with high fitness are selected to form the next generation population.

[0043] ⑧ Iterative optimization: Repeat the above selection, crossover, mutation and environment selection process until the preset number of iterations is reached or the convergence condition is met.

[0044] Iteration: The calculation of the motion response amplitude operator (RAO) and the process of adjusting the three parameters are repeatedly iterated until a design configuration that minimizes the motion response amplitude operator (RAO) to the target level is found, thereby obtaining an optimized design scheme.

[0045] The embodiments described above are merely illustrative of specific implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform, characterized in that, Includes the following steps: Select the basic layout; select the basic layout of the semi-submersible wind power generation platform. Initial parameters and hydrodynamic model: Determine the initial principal dimension parameters of the semi-submersible wind power platform and construct the hydrodynamic model; Filter key design variables, select key parameters from the main dimension parameters for subsequent optimization, and determine the design range of each key parameter; Sobol global sensitivity analysis uses the Sobol global sensitivity analysis method to perform sensitivity analysis on selected key parameters, quantifies the influence of each key parameter, and selects the N key parameters with the highest sensitivity. Narrow down the design scope and filter out designs that do not meet the basic hydrostatic requirements; Using frequency domain potential flow theory, the objective function is to minimize the response amplitude operator of six degrees of freedom. The minimization of the response amplitude operator includes maximizing the heave period, minimizing the heave motion response, and minimizing the pitch motion response. The NSGA-III algorithm is used to adjust N key parameters in a coordinated manner and iterate repeatedly until a design configuration that can minimize the response amplitude operator to the target level is found, thereby obtaining an optimized design scheme.

2. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 1, characterized in that, Based on the environmental data of the target wind field and the parameters of the selected wind turbine, the environmental data of the target wind field includes wave height, wave period and wind speed, and the parameters of the wind turbine include weight, hub height and thrust.

3. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 1, characterized in that, The hydrodynamic model was constructed using industry-standard frequency domain potential flow software.

4. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 1, characterized in that, The key parameters in the main dimensional parameters include column diameter Dc, pontoon height Hp, pontoon width Bp, platform draft T, and column spacing Ls.

5. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power platform according to claim 4, characterized in that, The N key parameters with the highest sensitivity were selected, including column diameter Dc, platform draft T, and pontoon height Hp.

6. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power platform according to claim 5, characterized in that, The NSGA-III algorithm is used to adjust N key parameters in a coordinated manner, including: Parameter initialization: First, determine the initial value range of the column diameter Dc, platform draft T, and buoy height Hp. These parameters serve as optimization variables for the NSGA-III algorithm. Population initialization: Randomly generate an initial population within the parameter value range, with each individual representing a set of parameter combinations; Fitness assessment: The fitness of each individual is assessed according to the objective function. The heave period, heave motion response and pitch motion response under each parameter combination are calculated to determine the fitness value of the individual. Selection operation: A selection mechanism based on non-dominated sorting is adopted to divide the population into different levels and prioritize the selection of non-dominated individuals into the next generation; Crossover operation: Perform crossover operation on selected individuals to generate new individuals, and explore new solution spaces by exchanging some gene information; Mutation operation: Perform mutation operation on newly generated individuals to change some of the individual's gene values ​​with a certain probability; Environmental selection: The new individuals resulting from crossover and mutation are merged with the parent population, and non-dominated sorting and crowding calculations are performed again. Individuals with high fitness are selected to form the next generation population. Iterative optimization: Repeat the above selection, crossover, mutation, and environment selection process until the preset number of iterations is reached or the convergence condition is met.

7. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 6, characterized in that, The crossover operation employs simulated binary crossover; the mutation operation employs methods such as polynomial mutation to increase population diversity and prevent the algorithm from getting trapped in local optima.

8. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 1, characterized in that, The design range of the key parameters is defined based on the actual lower and lower limits of manufacturing constraints, structural requirements, and engineering experience.

9. The design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 1, characterized in that, The narrowing of the design scope involves performing hydrostatic calculations on a series of promising designs identified from the Sobol global sensitivity analysis, and filtering out designs that do not meet the basic hydrostatic requirements.

10. A design method for optimizing the hydrodynamic performance of a semi-submersible wind power generation platform according to claim 9, characterized in that, The narrowing of the design scope also includes checking stability, freeboard, draft, and displacement.

Citation Information

Patent Citations

  • Parameter design method of floating type wind turbine generator semi-submersible platform

    CN117744515A

  • Main size design method for offshore wind power floating platform

    CN117763649A