Method for optimizing main dimensions of floating wind-solar-wave multi-energy coupling platform

Abstract

The application discloses a main dimension optimization method of a floating wind-solar-wave multi-energy coupling platform, belongs to the field of offshore floating wind power platform optimization, and is used for solving the problem that the existing optimization method ignores the influence of a wave energy device and a photovoltaic device on stability, and the main point is that an initial population is generated in a parameter space, and each individual e corresponds to a main dimension parameter d of the platform; wherein the main dimension parameter d comprises a floating foundation parameter, a photovoltaic panel parameter and a wave energy device parameter; according to the gradient of the average objective function of each target subpopulation and a resource exchange matrix and the gradient of an adaptive penalty function, a corresponding mutation operation is performed on each individual e by the target subpopulation to which the individual e belongs, so as to generate a main dimension parameter d' corresponding to the main dimension parameter d of the individual e; wherein the individual e belongs to different target subpopulations, and the elements in the resource exchange matrix used for performing the mutation operation on the individual e and the gradient of the average objective function of the target subpopulation are different.

Description

Technical Field

[0001] This invention belongs to the field of offshore floating wind power platform optimization, specifically involving a master-scale optimization method for a floating wind-solar-wave multi-energy coupled platform. Background Technology

[0002] Currently, floating platforms, through flexible anchoring, overcome seabed geological constraints, becoming a feasible solution for large-scale deep-sea development. Simultaneously, multi-energy complementarity in the ocean can significantly improve energy capture efficiency. Domestic research also exists on the development and master-scale optimization of floating wind-solar-wave multi-energy coupled platforms, but these studies all have certain problems. For example, current platform and master-scale optimization methods often focus on optimizing a single energy source or a single performance index; wind power floating body optimization prioritizes minimizing displacement, but neglects the impact of wave energy devices and photovoltaic devices on stability. Isolated control strategies exacerbate contradictions, leading to low efficiency in multi-energy synergy on the platform. Summary of the Invention

[0003] To overcome the aforementioned problems in the prior art, a master-scale optimization method for a floating wind-solar-wave multi-energy coupled platform, according to some embodiments of this application, includes...

[0004] S1. Generate an initial population in the parameter space. The initial population includes N individuals e, and each individual e corresponds to a principal scale parameter d of the platform. The principal scale parameter d includes floating foundation parameters, photovoltaic panel parameters, and wave energy device parameters.

[0005] S2. Use the control function system to calculate the intermediate parameters of the principal scale parameter d for any individual e in the current population;

[0006] S3. Based on the intermediate parameters, use a liquid neural network (LNN) to calculate the value of the objective function, the gradient of the objective function, the value of the constraint function, and the gradient of the constraint function for any individual e, including the principal scale parameter d.

[0007] The objective function includes the economic objective function. Energy efficiency objective function and the objective function of motion performance The constraints include motion performance constraints, energy efficiency constraints, stability constraints, and nonnegativity constraints.

[0008] S4. Based on the current iteration number t and the violation amount of the principal scale parameter d of any individual e for each constraint, calculate the adaptive penalty function of the principal scale parameter d of any individual e. and the gradient of the adaptive penalty function The amount of constraint violation is calculated based on the function value of the constraint;

[0009] S5. Objective function based on the principal scale parameter d of individual e. , and The value of is used to assign a target subgroup to any individual e. The target subgroup includes the producer subgroup, the converter subgroup, and the stabilizer subgroup.

[0010] S6. Based on the gradients of the objective functions of the principal scale parameter d for each individual e in the target subgroup, calculate the gradient of the average objective function of each target subgroup. The gradient of the average objective function of the target subgroup includes the gradient of the average objective function of the producer subgroup. The gradient of the average objective function of the converter subgroup The gradient of the average objective function of the stable subgroup ;Calculate the resource exchange matrix R based on the gradient of the average objective function of each target subgroup;

[0011] S7. Gradient of the average objective function and resource exchange matrix of each objective subgroup and the gradient of the adaptive penalty function Based on the target subgroup to which individual e belongs, perform the corresponding mutation operation on any individual e to generate the principal scale parameter d′ corresponding to the principal scale parameter d of individual e; where, individual e belongs to different target subgroups, and the resource exchange matrix used to perform the mutation operation on them is... The gradients of the elements in the target group and the average objective function of the target subgroup are different;

[0012] S8. Calculate the intermediate parameters of the principal scale parameter d′ for any individual e using the control function system;

[0013] S9. Based on the intermediate parameters, use a liquid neural network (LNN) to calculate the objective function value, gradient of the objective function, constraint function value, and gradient of the constraint function for any individual e, including the principal scale parameter d′.

[0014] Based on the current iteration number t and the amount of violation of each constraint by the principal scale parameter d′ of any individual e, calculate the adaptive penalty function of the principal scale parameter d′ of any individual e;

[0015] S10. Population renewal and selection, including

[0016] Individual e with principal scale parameter d and individual e with principal scale parameter d′ are merged to obtain a merged population, and the principal scale parameter in the merged population is represented as d.

[0017] Based on the objective function of the principal scale parameter d of individual e in the merged population , and and adaptive penalty function Calculate the corrected objective function value of the principal scale parameter d for any individual e in the merged population. ;

[0018] Based on the corrected objective function value Pareto sorting is performed on individual e in the merged population, and crowding distance between individuals at the same Pareto level is calculated.

[0019] Sort individuals e according to Pareto level from high to low and the squeezing distance within the same Pareto level from large to small. Select the top N individuals e in the merged population to form the next generation population, and use the next generation population as the current population.

[0020] S11. Iteratively execute steps S2-S10 until any of the following convergence conditions are met:

[0021] (1) For several consecutive generations, the relative rate of change of the HV index corresponding to the Pareto optimal solution set is less than the threshold. The HV index is the volume of the hypercube bounded by the Pareto front and a defined reference point;

[0022] (2) The preset maximum number of iterations T_max is reached;

[0023] When the convergence condition is met, the Pareto optimal solution set in the current population is output. The Pareto optimal solution set is the optimization result of the platform's principal scale parameters.

[0024] According to the principal scale optimization method for the floating wind-solar-wave multi-energy coupled platform in some embodiments of this application, the principal scale parameter d is represented by the following formula:

[0025]

[0026] In the formula, Indicates the diameter of the main floating body, Indicates the platform's draft. Indicates the column spacing. This represents the total area of ​​the photovoltaic panels. Indicates platform type depth, Indicates the power density of wave energy devices. Indicates the height of the lower floating body, This indicates the cross-sectional area of ​​the lower floating body.

[0027] According to the master-scale optimization method for floating wind-solar-wave multi-energy coupled platforms in some embodiments of this application, the objective function is... It can be expressed by the following formula:

[0028]

[0029] Among them, the economic objective function It can be expressed by the following formula:

[0030]

[0031] In the formula, This represents the present value of the platform's total lifecycle cost. Indicates annual power generation;

[0032] in, It can be expressed by the following formula:

[0033]

[0034] In the formula, This indicates the platform's initial investment cost. This represents the operating and maintenance cost in year t. This indicates the additional cost of wave energy devices. Indicates the discount rate. Indicates the platform's design life;

[0035] in, It can be expressed by the following formula:

[0036]

[0037] In the formula, , , , Indicates the cost coefficient. Indicates the diameter of the main floating body, This represents the total area of ​​the photovoltaic panels. Indicates the platform's draft. Indicates the cross-sectional area of ​​the lower floating body. Indicates the height of the lower floating body;

[0038] in, It can be expressed by the following formula:

[0039]

[0040] In the formula, This represents the cost coefficient of wave energy devices. Indicates the power density of wave energy devices. Indicates the cross-sectional area of ​​the lower floating body;

[0041] in, It can be expressed by the following formula:

[0042]

[0043] In the formula, Indicates the efficiency of the fan. Indicates the rated power of the fan. Indicates photovoltaic efficiency. Indicates the average annual irradiance. This represents the total area of ​​the photovoltaic panels. This indicates the wave energy conversion efficiency. Indicates the power density of wave energy devices. Indicates the length of the wave energy device;

[0044] Among them, the energy efficiency objective function It can be expressed by the following formula:

[0045]

[0046] In the formula, This represents the total annual power generation. This represents the theoretical maximum capture energy;

[0047] in, It can be expressed by the following formula:

[0048]

[0049] In the formula, Indicates the rated power of the fan. Indicates the average annual irradiance. This represents the total area of ​​the photovoltaic panels. Indicates the power density of wave energy devices. Indicates the length of the wave energy device;

[0050] Among them, the objective function of motion performance It can be expressed by the following formula:

[0051]

[0052] In the formula, This indicates the maximum pitch angle during a 50-year sea state event. This indicates the maximum roll angle during a 50-year sea state event. This indicates the maximum heave displacement during a 50-year sea state event.

[0053] According to the master-scale optimization method for a floating wind-solar-wave multi-energy coupled platform in some embodiments of this application, the motion performance constraint is expressed by the following formula:

[0054]

[0055] In the formula, Indicates the pitch angle. Indicates the roll angle. Indicates heave displacement;

[0056] The energy efficiency constraint is expressed by the following formula:

[0057]

[0058] In the formula, This indicates the wave energy conversion efficiency. Indicates the resonant frequency. Indicates the incident frequency of the wave;

[0059] The stability constraint is expressed by the following equation:

[0060]

[0061] In the formula, Indicates high longitudinal stability. This indicates high lateral stability;

[0062] The nonnegativity constraint is expressed by the following equation:

[0063]

[0064] In the formula, This represents the principal scale parameter.

[0065] According to the principal-scale optimization method of the floating wind-solar-wave multi-energy coupled platform in some embodiments of this application, wherein the liquid neural network (LNN) differential equation describes:

[0066]

[0067] In the formula, Indicates the activation state of the i-th neuron. Indicates time, Represents the liquid time constant. Indicates the neuron connection weights. Denotes a differentiable activation function. Indicates the relationship with the first The first neuron connected The activation state of neurons Indicates neuron bias terms. This represents the input stimulus for the design variable d.

[0068] According to the master-scale optimization method of the floating wind-solar-wave multi-energy coupled platform in some embodiments of this application, wherein the adaptive penalty function It can be expressed by the following formula:

[0069]

[0070] In the formula, Indicates the number of constraints. Represents the first individual d The amount of a constraint violation, if the constraint is satisfied. ;

[0071] in, Indicates the first The weight of the next iteration is expressed by the following formula:

[0072]

[0073] In the formula, Indicates the current iteration number. Indicates the total number of iterations. Indicates the initial weights;

[0074] Wherein, the gradient of the adaptive penalty function It can be expressed by the following formula:

[0075]

[0076] In the formula, This represents the gradient of the constraint function.

[0077] According to the master-scale optimization method for floating wind-solar-wave multi-energy coupled platforms in some embodiments of this application, the resource exchange matrix... It can be expressed by the following formula:

[0078]

[0079] In the formula, This represents the cost-optimized transfer rate from producer to converter. This represents the energy efficiency transfer rate from converter to stabilizer. This represents the motion stability transfer rate when a stable entity transforms into a producer.

[0080] in, It can be expressed by the following formula:

[0081]

[0082] in, It can be expressed by the following formula:

[0083]

[0084] in, It can be expressed by the following formula:

[0085]

[0086] In the formula, The gradient of the average objective function of the producer subgroup is represented by the following expression: The gradient of the average objective function of the subgroup of converters is given. This represents the gradient of the average objective function of the stable subgroup.

[0087] According to the master-scale optimization method for a floating wind-solar-wave multi-energy coupled platform in some embodiments of this application, the method involves performing a corresponding mutation operation on any individual e according to the target subgroup to which individual e belongs, to generate a master-scale parameter d′ corresponding to the master-scale parameter d of individual e, including...

[0088] The principal scale parameter d′ corresponding to the principal scale parameter d of individual e in the producer subgroup is expressed by the following formula:

[0089]

[0090] The principal scale parameter d′ corresponding to the principal scale parameter d of individual e in the transformer subgroup is expressed by the following formula:

[0091]

[0092] The principal scale parameter d′ corresponding to the principal scale parameter d of an individual e in the stable subgroup is expressed by the following formula:

[0093]

[0094] In the formula, η1, η2, and η3 represent the learning rates of the producer subgroup, the converter subgroup, and the stabilizer subgroup, respectively, and β represents the penalty gradient weight coefficient.

[0095] According to the master-scale optimization method for floating wind-solar-wave multi-energy coupled platforms in some embodiments of this application, the objective function value is modified. It can be expressed by the following formula:

[0096]

[0097] In the formula, This represents the adaptive penalty function.

[0098] According to the master-scale optimization method for a floating wind-solar-wave multi-energy coupled platform in some embodiments of this application, the control function system includes:

[0099] The roll angle control function is expressed by the following formula:

[0100]

[0101] In the formula, Indicates the roll angle. Indicates the sea state coefficient. Indicates the significant wave height. Represents gravitational acceleration. This indicates high longitudinal stability;

[0102] The pitch angle control function is expressed by the following formula:

[0103]

[0104] In the formula, Indicates the pitch angle. Indicates the sea state coefficient. Indicates the significant wave height. Represents gravitational acceleration. This indicates high lateral stability;

[0105] The heave displacement control function is expressed by the following equation:

[0106]

[0107] In the formula, express, Indicates the sea state coefficient. Indicates the period of the spectral peak. Indicates the significant wave height. Indicates the cross-sectional area of ​​the lower floating body. Indicates the diameter of the main floating body;

[0108] The longitudinal stability control function is expressed by the following formula:

[0109]

[0110] In the formula, Indicates high longitudinal stability. Indicates the diameter of the main floating body, Indicates the volume of water discharged. Indicates the cross-sectional area of ​​the lower floating body. Indicates the column spacing. Indicates the height of the center of gravity;

[0111] The wave energy efficiency function is expressed by the following formula:

[0112]

[0113] In the formula, This indicates the wave energy conversion efficiency. Indicates platform type depth, Indicates the diameter of the main floating body, Indicates the power density of the wave energy device;

[0114] The resonant frequency control function is expressed by the following equation:

[0115]

[0116] In the formula, Indicates the resonant frequency. Indicates the density of seawater. Indicates the cross-sectional area of ​​the lower floating body. Indicates the volume of water discharged. Indicates additional mass.

[0117] Beneficial effects:

[0118] This invention employs a master-scale optimization approach, setting platform parameters, photovoltaic power generation parameters, and wave energy generation parameters within the master-scale parameters, while considering the impact of wave energy devices and photovoltaic devices on platform stability. By leveraging the dynamic learning capabilities and symbiotic evolution mechanism of a liquid neural network, it addresses the multi-objective optimization challenge of floating multi-energy coupled platforms.

[0119] This invention optimizes the platform's main scale to improve its power generation efficiency and increase its economic viability by ensuring the platform's motion performance and survivability at a certain scale. This invention selects three objective functions—economic viability, energy efficiency, and motion performance—as the core of global optimization closely centered on the platform's main scale. The economic viability objective directly determines the construction cost by the main scale, ensuring the platform's commercial viability throughout its lifecycle. The energy efficiency objective relates to the comprehensive impact of the main scale on wind, solar, and wave energy capture capabilities, pursuing synergistic gains from multi-energy complementarity. The motion performance objective constrains the main scale to ensure the platform's stability and safety in harsh sea conditions, forming the foundation for the continuous and reliable operation of all energy equipment. These three objectives together constitute a complete closed-loop optimization of the platform's main scale in terms of "cost-benefit-survivability."

[0120] Economic objective function of this invention The cost and structural efficiency of the platform's main dimensions are determined by factors such as the diameter of the main float, draft, and height of the lower float, which directly affect the platform's structural weight, material usage, and construction costs. The photovoltaic area and wave energy device power density are related to equipment investment and power generation revenue. The optimization objective is to minimize the unit power generation cost by rationally configuring the main dimensions, while ensuring structural strength and stability, thereby achieving the platform's economic feasibility throughout its entire lifecycle. Economic objectives drive the design of main dimensions towards lightweighting, high power generation efficiency, and low cost, avoiding over-design or material waste.

[0121] Energy efficiency objective function of this invention This reflects the comprehensive impact of the main scale on multi-energy capture capabilities. Specifically, the column spacing and main float diameter affect the layout length of the wave energy device and the wave diffraction effect, thus influencing wave energy capture efficiency. The photovoltaic area directly determines the solar power generation potential. The float depth is closely related to wave energy conversion efficiency, reflecting the structure's influence on wave response. The energy efficiency objective is to maximize comprehensive energy utilization efficiency and improve power generation capacity per unit area / volume by optimizing the main scale combination. The energy efficiency objective guides the main scale design to achieve three-dimensional, high-efficiency capture in terms of spatial layout, structural form, and energy device coupling methods.

[0122] The motion performance objective function of this invention Ensuring the platform's safety and stability under operational conditions and extreme sea states is crucial. The diameter of the main float, the cross-sectional area of ​​the lower float, and the spacing between the columns directly affect the platform's stability, thus controlling its roll and pitch responses. Draft influences the natural heave period, preventing resonance with common wave periods. The kinematic performance objective is to suppress the platform's kinematic response in extreme sea states through principal scale optimization, ensuring the normal operation and structural safety of equipment such as wind turbines and photovoltaic systems. The kinematic performance objective constrains the principal scale design to achieve a balance between stability, kinematic response, and structural safety, avoiding sacrificing the platform's survivability in harsh sea states for excessive pursuit of power generation or economic efficiency.

[0123] This invention addresses the potential for designs that, while theoretically superior, are practically infeasible during optimization, such as those exhibiting insufficient stability or excessive motion response. The constraint handling mechanism enforces the satisfaction of constraints, eliminating infeasible solutions and ensuring the optimization results have practical engineering value. The motion performance constraints, energy efficiency constraints, stability constraints, and non-negativity constraints of this invention are coupled. For example, motion performance constraints are related to stability constraints, affecting the platform's survivability in extreme sea conditions. Energy efficiency is related to resonant frequency constraints, ensuring the efficient operation of wave energy devices. The constraint handling mechanism integrates these complex constraints into the optimization process, avoiding conflicts between objective functions and achieving multi-objective collaborative optimization.

[0124] The optimization method of this invention adopts an adaptive penalty function method, in which the weights are dynamically adjusted with the number of iterations. This allows the optimization algorithm to explore a wider design space in the early stage and strengthen the penalty in the later stage to converge to a feasible solution, thereby accelerating convergence and improving optimization accuracy.

[0125] In this invention, during the mutation and resource exchange process of MOSEA, the penalty function influences the behavior of the agents: Producers focus on economic efficiency; if the main scale variable leads to a cost reduction but violates constraints, the penalty term will increase, prompting producers to adjust variables. Converters focus on energy efficiency; the penalty term ensures that energy efficiency optimization does not sacrifice motion performance. Stabilizers focus on motion performance; the penalty term strengthens stability constraints, preventing platform instability. The resource exchange matrix R promotes gradient information sharing, helping agents collaborate to reduce constraint violations. The main function of the resource exchange matrix R is to promote dynamic balance and collaborative optimization among the three objective functions. Through the resource exchange matrix R, the three subgroups share gradient information during the mutation process, thereby avoiding objective conflicts, improving convergence speed, and ensuring that the optimization result is overall optimal in terms of cost, benefit, and survivability.

[0126] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0127] Figure 1This is a floating wind-solar-wave multi-energy coupling platform in an embodiment of the present invention.

[0128] Figure 2 This is a multi-objective symbiotic evolution logic diagram in an embodiment of the present invention.

[0129] Figure 3 This is a flowchart of the scale optimization operation for multi-objective symbiosis in an embodiment of the present invention.

[0130] In the picture:

[0131] 1. Floating foundation, 2. Wind turbine, 3. Wave energy device, 4. Energy management system, 5. Mooring and anchoring system, 6. Photovoltaic system, 11. Bottom float, 12. Column, 13. Horizontal brace, 14. Tower foundation. Detailed Implementation

[0132] The embodiments of this application are described in detail below with reference to the accompanying drawings. The principal scale refers to the platform's basic geometric parameters, etc. Principal scale optimization is a multi-objective, multi-constraint, and nonlinear engineering system optimization problem. The principal scale optimization method for the floating wind-solar-wave multi-energy coupled platform of this invention differs from existing wind power floating body optimization methods that aim to minimize displacement and do not include or ignore the influence of wave energy devices and photovoltaic devices on platform stability. Therefore, this invention sets platform parameters, photovoltaic power generation parameters, and wave energy power generation parameters in the principal scale parameters, considering the influence of wave energy devices and photovoltaic devices on platform stability.

[0133] like Figure 1 As shown, the floating wind-solar-wave multi-energy coupling platform of the present invention includes a floating foundation, a wave energy device, and a photovoltaic system, wherein the wave energy device and the photovoltaic system are mounted on the floating foundation. Preferably, the floating foundation includes a first column, a second column, a third column (tower foundation), cross braces, a truss, a lower floating body, and a mooring and anchoring system. The tower foundation supports the floating wind turbine, and the first column, the second column, and the tower foundation are distributed at the vertices of a triangle and mounted on the lower floating body. The first column is connected to the second column, the second column to the tower foundation, and the tower foundation to the first column via cross braces, which are distributed at the sides of the triangle. The truss is supported by the cross braces and positioned within the triangular area, and the photovoltaic system is mounted on the truss. The wave energy device is mounted on the columns and / or the tower foundation. The mooring and anchoring system is connected to the lower floating body.

[0134] The floating wind-solar-wave multi-energy coupling platform of this invention couples wind power, wave power, and photovoltaic power generation in a three-dimensional space. It features intelligent design to address the challenges of integrating multiple energy sources. The floating foundation is an equilateral triangle, with the columns, tower foundation, and lower floating body internally divided into multiple compartments to ensure platform stability and ballast. The bottom of the floating foundation is a rounded-corner equilateral triangle, with an embedded triangle cut out from within. This embedded triangle is then beveled at right angles to reduce stress concentration. Three cylindrical columns are placed concentrically at the rounded corners on the exterior of the lower floating body. To ensure structural continuity and constructability, the tangential connection between the cylindrical columns and the lower floating body is an arc-shaped connection structure, ensuring consistent structural connections and ease of construction.

[0135] Existing floating platforms typically place wave energy devices in the center of the platform or on independent floating bodies, which is prone to interference with wind turbine foundations and reduces wave capture efficiency. In this invention, a wave energy device consisting of multiple sets of adaptive hinged oscillating floats is installed on the outer facade of the platform's columns. The wave energy device's vertical shaft is vertically fixed to the platform's wave-facing surface via a bracket. The floats are bolted to bushings, which are fitted onto the vertical shaft. Shaft supports are located at both ends of the vertical shaft, and these supports are bolted to the support structure on the column's outer shell. Hydraulic cylinders are bolted to the support structure. The hydraulic system, serving as the secondary and tertiary energy conversion system for the wave energy device, is installed within the column's nacelle. The electrical system provides voltage regulation and energy storage, and monitors and displays the system components.

[0136] This invention utilizes the aforementioned wave energy arrangement scheme, employing a column as a rigid barrier. This causes incident waves to undergo diffraction and reflection superposition on the column's wave-facing surface, forming a local wave height amplification zone. The vortex acceleration zone created by the column's turbulence enhances the float's motion response. The float's hinged support is directly welded to the column's outer shell, eliminating the need for an independent support frame. When encountering large waves, the float is pushed against the column by the wave force, avoiding the swaying and collision risks of traditional independent floats. All hydraulic lines are radially arranged inside the column, allowing maintenance personnel to access them vertically from the deck through the top hatch of the column. The wave energy device reduces the platform's sway response and improves the stability of wind turbine power generation.

[0137] The platform's columns are connected by a cross bracing structure. This cross bracing structure also serves as the base for the photovoltaic support structure. A truss structure is then erected on this base to support the photovoltaic panels. This arrangement makes full use of the platform's three-dimensional space. Specifically, the supports create a double-layered power generation plane within the limited deck area. Raising the supports forms a bottom ventilation corridor, increasing wind speed and reducing module temperature by ≥12℃ (in summer). Wave guide ramps with a 30° inclination are installed at the edges of the photovoltaic array to reduce salt spray deposition.

[0138] In this invention, the third column is selected for mounting the wind turbine tower and turbine unit, i.e., the tower foundation. The other two columns only provide stabilizing support, resulting in a significant reduction in the total steel consumption of the platform compared to the four-column design. The wind turbine tower base is connected to the column via cross-shaped ribs that extend to the column's bulkhead at all four ends, enabling a shared maintenance shaft. The shaft's sidewall integrates wave energy hydraulic pipelines and wind turbine cables. The tower base achieves internal segmentation of the column's bulkhead, further mitigating the risk of stability loss in the event of a breach in the column's outer plate.

[0139] This invention addresses the multi-objective optimization challenge of floating multi-energy coupled platforms by leveraging the dynamic learning capabilities and symbiotic evolution mechanism of liquid neural networks. For example... Figure 3 As shown, the main-scale optimization method for the floating wind-solar-wave multi-energy coupled platform disclosed herein includes:

[0140] S1. Generate an initial population in the parameter space. The population includes N individuals e, and each individual e corresponds to a platform principal scale parameter d. The principal scale parameter d includes floating foundation parameters, photovoltaic panel parameters, and wave energy device parameters.

[0141] In this invention, the impact of wave energy devices and photovoltaic devices on platform stability is considered, and the principal scale parameter d is expressed by the following formula:

[0142]

[0143] In the formula, This indicates the diameter of the main buoy (in meters). This indicates the platform's draft (in meters). Indicates the column spacing (in meters). The total area of ​​the photovoltaic panels (unit: m²) 2 ), This indicates the platform depth (in meters). This indicates the power density of the wave energy device (unit: kW / m). Indicates the height of the lower body (in meters). Indicates the cross-sectional area of ​​the lower floating body (unit: m²). 2 ).

[0144] S2. For any individual e in the current population, the principal scale parameter d is calculated using the control function system to calculate the intermediate parameters;

[0145] The control function system includes roll angle control function, pitch angle control function, heave displacement control function, longitudinal stability control function, wave energy efficiency function, and resonance frequency control function.

[0146] The roll angle control function is expressed by the following formula:

[0147]

[0148] In the formula, Indicates the roll angle. Indicates the sea state coefficient. Indicates the significant wave height. Represents gravitational acceleration. This indicates high longitudinal stability.

[0149] The pitch angle control function is expressed by the following formula:

[0150]

[0151] In the formula, Indicates the pitch angle. Indicates the sea state coefficient. Indicates the significant wave height. Represents gravitational acceleration. This indicates high lateral stability.

[0152] The heave displacement control function is expressed by the following equation:

[0153]

[0154] In the formula, Indicates the amplitude of heave displacement. Indicates the sea state coefficient. Indicates the period of the spectral peak. Indicates the significant wave height. Indicates the cross-sectional area of ​​the lower floating body. Indicates the diameter of the main floating body.

[0155] The longitudinal stability control function is expressed by the following equation:

[0156]

[0157] In the formula, Indicates high longitudinal stability. Indicates the diameter of the main floating body, Indicates the volume of water discharged. Indicates the cross-sectional area of ​​the lower floating body. Indicates the column spacing. Indicates the height of the center of gravity.

[0158] The wave energy efficiency function is expressed by the following formula:

[0159]

[0160] In the formula, This indicates the wave energy conversion efficiency. Indicates platform type depth, Indicates the diameter of the main floating body, This indicates the power density of the wave energy device.

[0161] The resonant frequency control function is expressed by the following equation:

[0162]

[0163] In the formula, Indicates the resonant frequency. Indicates the density of seawater. Indicates the cross-sectional area of ​​the lower floating body. Indicates the volume of water discharged. Indicates additional mass.

[0164] S3. Based on the intermediate parameters, use a liquid neural network (LNN) to calculate the value of the objective function, the gradient of the objective function, the value of the constraint function, and the gradient of the constraint function for any individual e, including the principal scale parameter d.

[0165] The differential equation of the liquid neural network (LNN) describes:

[0166]

[0167] In the formula, Indicates the activation state of the i-th neuron. Indicates time, Represents the liquid time constant. Indicates the neuron connection weights. Denotes a differentiable activation function. express, Indicates neuron bias terms. This represents the input stimulus for the design variable d.

[0168] The objective function includes the economic objective function. Energy efficiency objective function and the objective function of motion performance The constraints include motion performance constraints, energy efficiency constraints, stability constraints, and non-negativity constraints.

[0169] The core objective of master-scale optimization is to improve the platform's power generation efficiency and increase its economic viability by ensuring its motion performance and survivability within a certain scale. This invention selects three objective functions—economic viability, energy efficiency, and motion performance—as the core of global optimization closely centered around the platform's master scale. The economic viability objective is directly determined by the master scale (such as the diameter of the floating body and the draft), ensuring the platform's commercial viability throughout its lifecycle. The energy efficiency objective relates to the comprehensive impact of the master scale (such as the spacing between columns and the photovoltaic area) on the ability to capture wind, solar, and wave energy, pursuing synergistic gains from multi-energy complementarity. The motion performance objective constrains the master scale (such as the cross-sectional area of ​​the floating body and the draft) to ensure the platform's stability and safety in harsh sea conditions, which is the foundation for the continuous and reliable operation of all energy equipment. Together, these three constitute a complete closed-loop optimization of the platform's master scale in terms of "cost-benefit-survivability."

[0170] Based on the above reasons, the objective function It can be expressed by the following formula:

[0171]

[0172] Among them, the economic objective function The cost and structural efficiency that determine the main dimensions of the platform are related to the diameter of the main buoy ( ), Draft ( ), lower floating body height ( Factors such as photovoltaic area directly affect the platform's structural weight, material usage, and construction cost. Wave energy device power density This relates to equipment investment and power generation revenue. The optimization objective is to minimize the unit power generation cost by rationally configuring the main scale, while satisfying structural strength and stability requirements, thereby achieving the platform's economic feasibility throughout its entire lifecycle. Economic objectives drive the design of the main scale towards lightweighting, high power generation efficiency, and low cost, avoiding over-design or material waste.

[0173] Based on the above reasons, the economic objective function It can be expressed by the following formula:

[0174]

[0175] In the formula, This represents the present value of the platform's total lifecycle cost. This indicates the annual power generation.

[0176] in, It can be expressed by the following formula:

[0177]

[0178] In the formula, This indicates the platform's initial investment cost. This represents the operating and maintenance cost in year t. This indicates the additional cost of wave energy devices. Indicates the discount rate. Indicates the platform's design lifespan.

[0179] in, It can be expressed by the following formula:

[0180]

[0181] In the formula, , , , Indicates the cost coefficient. Indicates the diameter of the main floating body, This represents the total area of ​​the photovoltaic panels. Indicates the platform's draft. Indicates the cross-sectional area of ​​the lower floating body. Indicates the height of the lower floating body.

[0182] in, It can be expressed by the following formula:

[0183]

[0184] In the formula, This represents the cost coefficient of wave energy devices. Indicates the power density of wave energy devices. This indicates the cross-sectional area of ​​the lower floating body.

[0185] in, It can be expressed by the following formula:

[0186]

[0187] In the formula, Indicates the efficiency of the fan. Indicates the rated power of the fan. Indicates photovoltaic efficiency. Indicates the average annual irradiance. This represents the total area of ​​the photovoltaic panels. This indicates the wave energy conversion efficiency. Indicates the power density of wave energy devices. Indicates the length of the wave energy device.

[0188] Unit description: (Dollar), (kWh), ($ / m 3 ), (Year), The value ranges from 0.35 to 0.45. (kW) The value should be between 0.15 and 0.22. (kWh / m 2 ).

[0189] Energy efficiency objective function This reflects the comprehensive impact of the main scale on multi-energy capture capability. Among these, the column spacing ( ), main floating body diameter ( The influence of wave energy device layout length ( The photovoltaic area (and wave diffraction effect) thus affect the wave energy capture efficiency. The depth of the floating body directly determines the potential for solar power generation. ) and wave energy conversion efficiency ( Closely related to the structure's response to waves, the energy efficiency objective aims to maximize overall energy utilization efficiency and increase power generation capacity per unit area / volume by optimizing the combination of master scales. This energy efficiency objective guides the master scale design to achieve three-dimensional, high-efficiency energy capture in terms of spatial layout, structural form, and energy device coupling methods.

[0190] Based on the above reasons, the energy efficiency objective function It can be expressed by the following formula:

[0191]

[0192] In the formula, This represents the total annual power generation. This represents the theoretical maximum capture energy.

[0193] in, It can be expressed by the following formula:

[0194]

[0195] In the formula, Indicates the rated power of the fan. Indicates the average annual irradiance. This represents the total area of ​​the photovoltaic panels. Indicates the power density of wave energy devices. Indicates the length of the wave energy device.

[0196] Unit description: (°), (°), (m).

[0197] Motion performance objective function Ensure the platform's safety and stability under operational conditions and extreme sea conditions. Specifically, the diameter of the main buoy ( ), cross-sectional area of ​​the lower floating body ( ), column spacing ( This directly affects the platform's stability. , ), thereby controlling the roll and pitch responses. Draft ( The inherent heave period is affected to avoid resonance with common wave periods. The kinematic performance objective is to suppress the platform's kinematic response under extreme sea conditions through principal-scale optimization, ensuring the normal operation and structural safety of equipment such as wind turbines and photovoltaic systems. The kinematic performance objective constrains the principal-scale design to achieve a balance between stability, kinematic response, and structural safety, avoiding sacrificing the platform's survivability in harsh sea conditions for excessive pursuit of power generation or economic efficiency.

[0198] Based on the above reasons, the objective function for motion performance... It can be expressed by the following formula:

[0199]

[0200] In the formula, This indicates the maximum pitch angle during a 50-year sea state event. This indicates the maximum roll angle during a 50-year sea state event. This indicates the maximum heave displacement during a 50-year sea state event.

[0201] Unit description: (m), (s), (m), (m), (m), (m) 3 ), (1025 kg / m 3 ), (kg) , , Values ​​range from 0.15 to 0.25 s 2 / m, (m), (m) 2 ), (m).

[0202] As described above, the constraints of this invention include motion performance constraints, energy efficiency constraints, stability constraints, and non-negativity constraints. The selection of these constraints takes into account:

[0203] (1) The constraints cover important aspects of platform design, including safety (motion performance and stability), efficiency (energy efficiency) and feasibility (non-negativity). They form a closed loop with the three objective functions (economy, energy efficiency, and motion performance) to ensure that the optimization results are balanced in terms of cost, benefits and survivability, thus achieving the multi-objective collaborative optimization requirements.

[0204] (2) The constraint values ​​are based on industry standards, experimental data and simulation experience to ensure that the platform meets the design requirements of offshore renewable energy projects and achieves engineering practice and standard compliance.

[0205] (3) Constraints are integrated into the multi-objective co-evolutionary algorithm (MOSEA) through an adaptive penalty function mechanism, dynamically adjusting the weights to help the algorithm avoid local optima when exploring the design space and converge to the practical solution in engineering, thereby achieving optimization algorithm guidance.

[0206] (4) Constraint processing prevents the risks of “over-optimization”, such as reducing the scale to reduce costs, resulting in insufficient stability, or ignoring motion response to improve energy efficiency, thus achieving risk control.

[0207] The optimization method and constraints of this invention not only improve the platform's economy and energy efficiency but also ensure its reliability, safety, and durability in real marine environments. Based on these factors, the complete constraint equations include kinematic performance constraints, energy efficiency constraints, stability constraints, and non-negativity constraints.

[0208] The main function of the motion performance constraints of this invention is to ensure the platform's survivability in harsh sea conditions, reduce downtime, and improve the lifespan and reliability of energy equipment; by constraining motion response, it avoids "high-performance but high-risk" designs in the optimization results, ensuring the feasibility of the platform in actual deployment; and it provides key boundary conditions for multi-objective optimization, ensuring that economic and energy efficiency objectives are not achieved at the expense of safety.

[0209] Key considerations include:

[0210] (1) Platform safety and equipment reliability: In deep-sea environments, platforms face extreme sea conditions (such as 50-year return period waves), and excessive motion responses (such as pitch, roll and heave) can lead to structural fatigue, equipment damage or reduced power generation efficiency. For example, wind turbine towers and blades are sensitive to motion angles, and pitch or roll exceeding 10° may cause wind turbine shutdown or mechanical failure; excessive heave displacement will affect the coupling efficiency between wave energy devices and waves.

[0211] (2) Industry standards and empirical values: These thresholds reference offshore floating platform design specifications (such as DNV-GL standards) and actual engineering cases to ensure that the platform maintains an acceptable range of motion under operating conditions and sea conditions. The constraint values ​​(10° and 5.0 m) are motion limits based on typical floating wind power platforms, balancing safety and economy.

[0212] (3) Coordination requirements of multi-energy coupling system: The platform integrates wind power, photovoltaic and wave energy devices, and the motion performance directly affects the energy capture efficiency. For example, photovoltaic panels need a relatively stable base to avoid cracks, and wave energy devices require the platform's heave response to match the wave period.

[0213] Based on the above reasons, the motion performance constraint is expressed by the following formula:

[0214]

[0215] In the formula, Indicates the pitch angle. Indicates the roll angle. This indicates heave displacement.

[0216] The main function of the energy efficiency constraints in this invention is to enhance the power generation potential of wave energy devices and ensure the maximization of the overall energy output of the multi-energy coupling system. Through frequency matching, the interaction between the platform and the waves is optimized, reducing energy loss and improving wave energy conversion efficiency. This prevents the optimization algorithm from excessively reducing the platform size (e.g., decreasing the size) in pursuit of economic efficiency. or This leads to low wave energy efficiency.

[0217] Key considerations include:

[0218] (1) Economic feasibility of wave energy devices: wave energy conversion efficiency ( The efficiency directly determines the revenue generated from power generation. When the efficiency is below 0.15, the power generation of the wave energy device may not cover the cost, affecting the overall economic viability of the platform. This threshold is based on the maturity of wave energy technology and historical data to ensure that the device has the minimum acceptable efficiency under typical sea conditions.

[0219] (2) Synergistic gain of multi-energy complementarity: The wave energy device is coupled with the platform structure (such as the column diffraction effect), and the efficiency constraint ensures that the contribution of wave energy is not ignored, thereby maximizing the total annual power generation. ).

[0220] Based on the above reasons, the energy efficiency constraint is expressed by the following formula:

[0221]

[0222] In the formula, This indicates the wave energy conversion efficiency. Indicates the resonant frequency. This indicates the incident frequency of the wave.

[0223] The primary function of the stability constraints in this invention is to ensure the platform maintains structural integrity and safety throughout its lifecycle, preventing overturning accidents. It provides the basis for motion performance constraints, as stability directly determines roll and pitch responses. It guides the optimization algorithm to pursue lightweight (economical) design without sacrificing fundamental safety metrics.

[0224] Key considerations include:

[0225] (1) Basic requirements for platform stability: Center height (GM) is a key parameter for measuring the platform's ability to resist capsizing. A GM value that is too low (e.g., <0.15 m) will result in insufficient restoring moment and make the platform prone to capsizing under wind and wave loads. This threshold is based on the design specifications for offshore floating platforms (such as IMO standards) to ensure that the platform has sufficient stability reserves under operation and extreme sea conditions.

[0226] (2) Structural safety and load distribution: GM value and platform principal dimensions (e.g.) , , Directly related to GM, the constraint ensures that the optimization results meet the hydrostatic stability requirements. For example, the cross-sectional area of ​​the lower floating body ( ) and column spacing ( )Influence and The calculation.

[0227] (3) Impact of multi-device integration: The wind turbine, photovoltaic array and wave energy device on the upper part of the platform increase the center of gravity height (KG). Stability constraints compensate for the impact of these additional masses and prevent unstable layouts with a "top-heavy" design from occurring in the optimization design.

[0228] Based on the above reasons, the stability constraint is expressed by the following equation:

[0229]

[0230] In the formula, Indicates high longitudinal stability. This indicates high lateral stability.

[0231] The main function of the nonnegative constraint in this invention is to ensure the engineering feasibility of the optimization results, and that all principal scale parameters can be practically applied. It simplifies the optimization process, reduces the number of infeasible solutions, and improves algorithm efficiency.

[0232] Key considerations include:

[0233] (1) Physical feasibility: Design variables (e.g.) , , The value represents the platform's physical dimensions (diameter, height, area), and must be positive to have engineering significance. Negative values ​​make construction or operation impossible in reality.

[0234] (2) Reasonableness of mathematical optimization: In the optimization algorithm, non-negativity constraints prevent the design variables from converging to the invalid domain and avoid numerical calculation errors.

[0235] Based on the above reasons, the nonnegativity constraint is expressed by the following equation:

[0236]

[0237] In the formula, This represents the principal scale parameter.

[0238] In this process, principal-scale optimization involves multiple design variables that directly impact the platform's economy, energy efficiency, and motion performance. However, the optimization process may result in designs that are theoretically superior but practically infeasible, such as insufficient stability or excessive motion response. The constraint handling mechanism ensures the optimization results have practical engineering value by forcibly satisfying constraints and eliminating infeasible solutions. The motion performance constraints, energy efficiency constraints, stability constraints, and non-negativity constraints of this invention are coupled. For example, motion performance constraints (such as pitch angle ≤ 10°) are related to stability constraints (such as center-of-gravity height ≥ 0.15m), affecting the platform's survivability in extreme sea conditions. Energy efficiency constraints (such as wave energy conversion efficiency ≥ 0.15) are related to resonant frequency constraints, ensuring the efficient operation of the wave energy device. The constraint handling mechanism integrates these complex constraints into the optimization process, avoiding conflicts between objective functions and achieving multi-objective collaborative optimization.

[0239] S4. Based on the current iteration number t and the violation amount of the principal scale parameter d of any individual e for each constraint, calculate the adaptive penalty function of the principal scale parameter d of any individual e. and the gradient of the adaptive penalty function The amount of constraint violation is calculated based on the function value of the constraint.

[0240] The adaptive penalty function is expressed by the following equation:

[0241]

[0242] In the formula, Indicates the number of constraints. Represents the first individual d The amount of constraint violation.

[0243] For example, if the constraint is Then the quantity is violated If the constraints are satisfied, then .

[0244] in, Indicates the first The weight of each iteration increases exponentially with the iteration number t, as expressed by the following formula:

[0245]

[0246] In the formula, Indicates the current iteration number. Indicates the total number of iterations. This represents the initial weights.

[0247] Wherein, the gradient of the adaptive penalty function It can be expressed by the following formula:

[0248]

[0249] In the formula, The gradient of the constraint function is represented. Indicates the number of constraints.

[0250] Optimization methods (such as genetic algorithms) may rely on penalty functions when handling constraints, but fixed weights may lead to premature convergence or failure to converge to the feasible region. This invention employs an adaptive penalty function method, where weights are dynamically adjusted with the number of iterations. This allows the optimization algorithm to explore a wider design space in the early stages (allowing for minor constraint violations), and then strengthens the penalty in the later stages to converge to a feasible solution, thereby accelerating convergence and improving optimization accuracy.

[0251] This invention uses the weighted sum of squares of constraint violations as a penalty term. And add it to the optimization objectives. In MOSEA, as... Figure 2 As shown, the three types of intelligent agents—producers, converters, and stabilizers—consider a penalty term when evaluating the objective function. For example, the actual objective function can be adjusted as follows:

[0252]

[0253] in This is the original multi-objective function. Thus, individuals that violate the constraints will be penalized, and their fitness will decrease.

[0254] During the iteration process, the penalty weight It grows exponentially with the number of iterations. Initial weights The initial weights are set relatively small, allowing the algorithm to explore infeasible regions (which may contain potential optimal solutions) early on; as iterations progress, the weights increase, forcing the algorithm to converge toward the feasible region. This mechanism balances global search and local convergence, achieving adaptive weight updates.

[0255] In the mutation and resource exchange processes of MOSEA, the penalty function influences the agent's behavior: Producers focus on economics; if the main scale variable leads to a cost reduction but violates constraints, the penalty term increases, prompting producers to adjust variables. Converters focus on energy efficiency; the penalty term ensures that energy efficiency optimization does not sacrifice motion performance. Stabilizers focus on motion performance; the penalty term strengthens stability constraints, preventing platform instability. The resource exchange matrix R promotes gradient information sharing, helping agents collaborate to reduce constraint violations.

[0256] S5. Objective function based on the principal scale parameter d of individual e. , and The value of is used to assign a target subgroup to any individual e. The target subgroup includes a producer subgroup, a converter subgroup, and a stabilizer subgroup.

[0257] Based on the value of the objective function of the principal scale parameter d of individual e, the population is divided into three objective subgroups:

[0258] Producer Subgroup: Economic Objective Function The optimal N / 3 individuals.

[0259] Converter subgroup: Energy efficiency objective function The optimal N / 3 individuals.

[0260] Stable subgroup: Motion performance objective function The optimal N / 3 individuals.

[0261] Based on the gradients of the objective functions of the principal scale parameter d for each individual e in the target subgroup, the gradient of the average objective function of each target subgroup is calculated. This average objective function gradient includes the gradient of the average objective function of the producer subgroup. The gradient of the average objective function of the converter subgroup The gradient of the average objective function of the stable subgroup The resource exchange matrix R is calculated based on the gradient of the average objective function of each objective subgroup.

[0262] The multi-objective symbiotic evolution mechanism (MOSEA) of this invention utilizes the synergistic action of three types of intelligent agents: producers, converters, and stabilizers. Producers focus on the economic objective function. The converter focuses on the energy efficiency objective function. The objective function for stabilizing motion performance .

[0263] The main function of the resource exchange matrix R is to promote dynamic balance and collaborative optimization among the three objective functions. Through the resource exchange matrix R, the three subgroups share gradient information during the mutation process, thereby avoiding objective conflicts, improving the convergence speed, and ensuring that the optimization results are optimal in terms of cost, benefit, and survivability.

[0264] Its effects are manifested as follows:

[0265] (1) Information transmission: The producer’s cost optimization information is transmitted to the converter, the converter’s energy efficiency information is transmitted to the stabilizer, and the stabilizer’s motion stability information is transmitted to the producer, forming a closed-loop optimization.

[0266] (2) Conflict resolution: By exchanging gradients, the conflict between objective functions (such as the trade-off between economy and motion performance) is alleviated, and the population is guided to evolve toward the Pareto optimal solution set.

[0267] (3) Dynamic adaptation: The elements of matrix R are dynamically updated during the iteration process to ensure that the optimization process adapts to the needs of different design stages.

[0268] Among them, the resource exchange matrix It can be expressed by the following formula:

[0269]

[0270] In the formula, This represents the cost-optimized transfer rate from producer to converter. This represents the energy efficiency transfer rate from converter to stabilizer. This represents the motion stability transfer rate when a stabilizer is converted into a producer.

[0271] Calculate the gradient of the producer-based objective function gradient of the objective function of the converter The directional consistency between them.

[0272] dot product Reflecting the consistency of gradient direction: positive values ​​indicate similar directions, encouraging information transmission; negative values ​​indicate conflict, reducing transmission. By taking non-negative values, the transmission rate is ensured to be non-negative.

[0273] in, It can be expressed by the following formula:

[0274]

[0275] Calculate the gradient of the objective function based on the converter. gradient of the objective function of the stabilizer The directional consistency between them, while introducing second-order information. This is to resolve the energy efficiency-stability conflict. In actual calculations, the second derivative is difficult to obtain directly, so the gradient dot product approximation is often used.

[0276] in, It can be expressed by the following formula:

[0277]

[0278] Calculate the gradient of the objective function based on the stabilizer. gradient of the producer's objective function Consistency in direction between them:

[0279] in, It can be expressed by the following formula:

[0280]

[0281] In the formula, The gradient of the average objective function of the producer subgroup is represented by the following expression: The gradient of the average objective function of the subgroup of converters is given. This represents the gradient of the average objective function of the stable subgroup.

[0282] S7. Gradient of the average objective function and resource exchange matrix of each objective subgroup and the gradient of the adaptive penalty function Based on the target subgroup to which individual e belongs, perform the corresponding mutation operation on any individual e to generate the principal scale parameter d′ corresponding to the principal scale parameter d of individual e; where, individual e belongs to different target subgroups, and the resource exchange matrix used to perform the mutation operation on them is... The gradients of the elements in the target group and the average objective function of the target subgroup are different.

[0283] Among them, the principal scale parameter d′ corresponding to the principal scale parameter d of individual e in the producer subgroup is expressed by the following formula:

[0284]

[0285] The principal scale parameter d′ corresponding to the principal scale parameter d of individual e in the transformer subgroup is expressed by the following formula:

[0286]

[0287] The principal scale parameter d′ corresponding to the principal scale parameter d of an individual e in the stable subgroup is expressed by the following formula:

[0288]

[0289] In the formula, η1, η2, and η3 represent the learning rates of the producer subgroup, the converter subgroup, and the stabilizer subgroup, respectively, and β represents the penalty gradient weight coefficient.

[0290] S8. For any individual e, calculate the intermediate parameters using the control function system.

[0291] S9. Based on the intermediate parameters, using a liquid neural network (LNN), calculate the objective function value, gradient of the objective function, constraint function value, and gradient of the constraint function for any individual e's principal scale parameter d′; based on the current iteration number t and the violation amount of individual e's principal scale parameter d′ for each constraint, calculate the adaptive penalty function for any individual e's principal scale parameter d′. It is understandable that and The calculation formulas are the same; only the variables differ.

[0292] S10. Population renewal and selection, including

[0293] Individual e with principal scale parameter d and individual e with principal scale parameter d′ are merged to obtain a merged population, and the principal scale parameter platform in the merged population is represented as d.

[0294] Based on the objective function of the principal scale parameter d of individual e , and and adaptive penalty function Calculate the corrected objective function value of the principal scale parameter d for any individual e in the merged population. .

[0295] Among them, the objective function value is corrected. It can be expressed by the following formula:

[0296]

[0297] In the formula, This represents the adaptive penalty function.

[0298] Among them, non-dominated sorting: based on the modified objective function value Pareto sorting is performed on individuals e in the merged population, and crowding distance between individuals at the same Pareto level is calculated.

[0299] In the elite selection process, individuals e are sorted from high to low according to their Pareto level and from large to small within the same Pareto level. The top N individuals e in the merged population are selected to form the next generation population, which is then used as the current population.

[0300] S11. Update iteration count: t = t + 1, that is, iterate through steps S2-S10 until any of the following convergence conditions are met:

[0301] (1) For 10 consecutive generations, the relative rate of change of the HV index corresponding to the Pareto optimal solution set is less than the threshold. ( =1%), where the HV index is the volume of the hypercube bounded by the Pareto front and a set reference point.

[0302] (2) The preset maximum number of iterations T_max is reached.

[0303] When the convergence condition is met, the Pareto optimal solution set in the current population is output. The Pareto optimal solution set is the optimization result of the platform's principal scale parameters.

[0304] The optimization results are shown in Tables 1 and 2, where Table 2 shows the optimization results of the improved objective function.

[0305] Table 1

[0306]

[0307] Table 2

[0308]

[0309] Based on the above embodiments, this application also provides a computer program that, when run on a computer, causes the computer to execute the methods provided in the above embodiments.

[0310] Based on the above embodiments, this application also provides a computer storage medium storing a computer program, which, when executed by a computer, causes the computer to perform the methods provided in the above embodiments.

[0311] The storage medium can be any available medium that a computer can access. For example, but not limited to, a computer-readable medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

[0312] Based on the above embodiments, this application also provides a chip for reading a computer program stored in a memory to implement the method provided in the above embodiments.

[0313] Based on the above embodiments, this application provides a computer program product that implements the methods provided in the above embodiments when the computer program product is run on an electronic device.

[0314] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0315] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0316] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0317] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0318] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A master-scale optimization method for a floating wind-solar-wave multi-energy coupled platform, characterized in that, include S1. Generate an initial population in the parameter space. The initial population includes N individuals e, and each individual e corresponds to a principal scale parameter d of the platform. The principal scale parameter d includes floating foundation parameters, photovoltaic panel parameters, and wave energy device parameters. S2. Use the control function system to calculate the intermediate parameters of the principal scale parameter d for any individual e in the current population; S3. Based on the intermediate parameters, use a liquid neural network (LNN) to calculate the value of the objective function, the gradient of the objective function, the value of the constraint function, and the gradient of the constraint function for any individual e, including the principal scale parameter d. The objective function includes the economic objective function. Energy efficiency objective function and the objective function of motion performance The constraints include motion performance constraints, energy efficiency constraints, stability constraints, and nonnegativity constraints. S4. Based on the current iteration number t and the violation amount of the principal scale parameter d of any individual e for each constraint, calculate the adaptive penalty function of the principal scale parameter d of any individual e. and the gradient of the adaptive penalty function The amount of constraint violation is calculated based on the function value of the constraint; S5. Objective function based on the principal scale parameter d of individual e. , and The value of is used to assign a target subgroup to any individual e. The target subgroup includes the producer subgroup, the converter subgroup, and the stabilizer subgroup. S6. Based on the gradients of the objective functions of the principal scale parameter d for each individual e in the target subgroup, calculate the gradient of the average objective function of each target subgroup. The gradient of the average objective function of the target subgroup includes the gradient of the average objective function of the producer subgroup. The gradient of the average objective function of the converter subgroup The gradient of the average objective function of the stable subgroup ;Calculate the resource exchange matrix R based on the gradient of the average objective function of each target subgroup; S7. Gradient of the average objective function and resource exchange matrix of each objective subgroup and the gradient of the adaptive penalty function Based on the target subgroup to which individual e belongs, perform the corresponding mutation operation on any individual e to generate the principal scale parameter d′ corresponding to the principal scale parameter d of individual e; where, individual e belongs to different target subgroups, and the resource exchange matrix used to perform the mutation operation on them is... The gradients of the elements in the target group and the average objective function of the target subgroup are different; S8. Calculate the intermediate parameters of the principal scale parameter d′ for any individual e using the control function system; S9. Based on the intermediate parameters, use a liquid neural network (LNN) to calculate the objective function value, gradient of the objective function, constraint function value, and gradient of the constraint function for any individual e, including the principal scale parameter d′. Based on the current iteration number t and the amount of violation of each constraint by the principal scale parameter d′ of any individual e, calculate the adaptive penalty function of the principal scale parameter d′ of any individual e; S10. Population renewal and selection, including Individual e with principal scale parameter d and individual e with principal scale parameter d′ are merged to obtain a merged population, and the principal scale parameter in the merged population is represented as d. Based on the objective function of the principal scale parameter d of individual e in the merged population , and and adaptive penalty function Calculate the corrected objective function value of the principal scale parameter d for any individual e in the merged population. ; Based on the corrected objective function value Pareto sorting is performed on individual e in the merged population, and crowding distance between individuals at the same Pareto level is calculated. Individuals e are sorted from high to low according to Pareto level and from large to small within the same Pareto level. The top N individuals e in the merged population are selected to form the next generation population, and the next generation population is used as the current population. S11. Iteratively execute steps S2-S10 until any of the following convergence conditions are met: (1) For several consecutive generations, the relative rate of change of the HV index corresponding to the Pareto optimal solution set is less than the threshold. The HV index is the volume of the hypercube bounded by the Pareto front and a defined reference point; (2) Reach the preset maximum number of iterations T_max; When the convergence condition is met, the Pareto optimal solution set in the current population is output. The Pareto optimal solution set is the optimization result of the platform's principal scale parameters.

2. The main-scale optimization method for the floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, The principal scale parameter d is expressed by the following formula: In the formula, Indicates the diameter of the main floating body, Indicates the platform's draft. Indicates the column spacing. This represents the total area of ​​the photovoltaic panels. Indicates platform type depth, Indicates the power density of wave energy devices. Indicates the height of the lower floating body, This indicates the cross-sectional area of ​​the lower floating body.

3. The main-scale optimization method for the floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, objective function It can be expressed by the following formula: Among them, the economic objective function It can be expressed by the following formula: In the formula, This represents the present value of the platform's total lifecycle cost. Indicates annual power generation; in, It can be expressed by the following formula: In the formula, This indicates the platform's initial investment cost. This represents the operating and maintenance cost in year t. This indicates the additional cost of wave energy devices. Indicates the discount rate. Indicates the platform's design life; in, It can be expressed by the following formula: In the formula, , , , Indicates the cost coefficient. Indicates the diameter of the main floating body, This represents the total area of ​​the photovoltaic panels. Indicates the platform's draft. Indicates the cross-sectional area of ​​the lower floating body. Indicates the height of the lower floating body; in, It can be expressed by the following formula: In the formula, This represents the cost coefficient of wave energy devices. Indicates the power density of wave energy devices. Indicates the cross-sectional area of ​​the lower floating body; in, It can be expressed by the following formula: In the formula, Indicates the efficiency of the fan. Indicates the rated power of the fan. Indicates photovoltaic efficiency. Indicates the average annual irradiance. This represents the total area of ​​the photovoltaic panels. This indicates the wave energy conversion efficiency. Indicates the power density of wave energy devices. Indicates the length of the wave energy device; Among them, the energy efficiency objective function It can be expressed by the following formula: In the formula, This represents the total annual power generation. This represents the theoretical maximum capture energy; in, It can be expressed by the following formula: In the formula, Indicates the rated power of the fan. Indicates the average annual irradiance. This represents the total area of ​​the photovoltaic panels. Indicates the power density of wave energy devices. Indicates the length of the wave energy device; Among them, the objective function of motion performance It can be expressed by the following formula: In the formula, This indicates the maximum pitch angle during a 50-year sea state event. This indicates the maximum roll angle during a 50-year sea state event. This indicates the maximum heave displacement during a 50-year sea state event.

4. The main-scale optimization method for the floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, Motion performance constraints are expressed by the following formula: In the formula, Indicates the pitch angle. Indicates the roll angle. Indicates heave displacement; The energy efficiency constraint is expressed by the following formula: In the formula, This indicates the wave energy conversion efficiency. Indicates the resonant frequency. Indicates the incident frequency of the wave; The stability constraint is expressed by the following equation: In the formula, Indicates high longitudinal stability. This indicates high lateral stability; The nonnegativity constraint is expressed by the following equation: In the formula, This represents the principal scale parameter.

5. The main-scale optimization method for a floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, Liquid Neural Network (LNN) Differential Equation Description: In the formula, Indicates the activation state of the i-th neuron. Indicates time, Represents the liquid time constant. Indicates the neuron connection weights. Denotes a differentiable activation function. Indicates the relationship with the first The first neuron connected The activation state of neurons Indicates neuron bias terms. This represents the input stimulus for the design variable d.

6. The main-scale optimization method for a floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, Adaptive penalty function It can be expressed by the following formula: In the formula, Indicates the number of constraints. Represents the first individual d The amount of a constraint violation, if the constraint is satisfied. ; in, Indicates the first The weight of the next iteration is expressed by the following formula: In the formula, Indicates the current iteration number. Indicates the total number of iterations. Indicates the initial weights; Wherein, the gradient of the adaptive penalty function It can be expressed by the following formula: In the formula, This represents the gradient of the constraint function.

7. The main-scale optimization method for a floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, Resource exchange matrix It can be expressed by the following formula: In the formula, This represents the cost-optimized transfer rate from producer to converter. This represents the energy efficiency transfer rate from converter to stabilizer. The motion stability transfer rate represents the transformation of a stabilizer into a producer. in, It can be expressed by the following formula: in, It can be expressed by the following formula: in, It can be expressed by the following formula: In the formula, The gradient of the average objective function of the producer subgroup is represented by the following expression: The gradient of the average objective function of the subgroup of converters is given. This represents the gradient of the average objective function of the stable subgroup.

8. The main-scale optimization method for a floating wind-solar-wave multi-energy coupled platform according to claim 7, characterized in that, in, Based on the target subgroup to which individual e belongs, perform the corresponding mutation operation on any individual e to generate the principal scale parameter d′ corresponding to the principal scale parameter d of individual e, including The principal scale parameter d′ corresponding to the principal scale parameter d of individual e in the producer subgroup is expressed by the following formula: The principal scale parameter d′ corresponding to the principal scale parameter d of individual e in the transformer subgroup is expressed by the following formula: The principal scale parameter d′ corresponding to the principal scale parameter d of an individual e in the stable subgroup is expressed by the following formula: In the formula, η1, η2, and η3 represent the learning rates of the producer subgroup, the converter subgroup, and the stabilizer subgroup, respectively, and β represents the penalty gradient weight coefficient.

9. The main-scale optimization method for a floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, Correct the objective function value It can be expressed by the following formula: In the formula, This represents the adaptive penalty function.

10. The main-scale optimization method for a floating wind-solar-wave multi-energy coupled platform according to claim 1, characterized in that, in, The control function system includes: The roll angle control function is expressed by the following formula: In the formula, Indicates the roll angle. Indicates the sea state coefficient. Indicates the significant wave height. Represents gravitational acceleration. This indicates high longitudinal stability; The pitch angle control function is expressed by the following formula: In the formula, Indicates the pitch angle. Indicates the sea state coefficient. Indicates the significant wave height. Represents gravitational acceleration. This indicates high lateral stability; The heave displacement control function is expressed by the following equation: In the formula, express, Indicates the sea state coefficient. Indicates the period of the spectral peak. Indicates the significant wave height. Indicates the cross-sectional area of ​​the lower floating body. Indicates the diameter of the main floating body; The longitudinal stability control function is expressed by the following formula: In the formula, Indicates high longitudinal stability. Indicates the diameter of the main floating body, Indicates the volume of water discharged. Indicates the cross-sectional area of ​​the lower floating body. Indicates the column spacing. Indicates the height of the center of gravity; The wave energy efficiency function is expressed by the following formula: In the formula, This indicates the wave energy conversion efficiency. Indicates platform type depth, Indicates the diameter of the main floating body, Indicates the power density of the wave energy device; The resonant frequency control function is expressed by the following equation: In the formula, Indicates the resonant frequency. Indicates the density of seawater. Indicates the cross-sectional area of ​​the lower floating body. Indicates the volume of water discharged. Indicates additional mass.

Images