Flexible DC power transmission optimization control method suitable for offshore wind power integration

Abstract

The invention discloses a flexible DC power transmission optimization control method suitable for offshore wind power integration, and relates to the technical field of offshore power transmission control, and the method comprises the steps: collecting the latitude and longitude coordinates and offshore distance of each wind power plant, and collecting the environmental parameters of each wind power plant; setting a sea surface environment complexity prediction strategy to obtain a marine environment complexity index; calculating the relative position and interdependence of each wind power plant in the whole area according to the spatial distribution of the wind power plants, obtaining the position of a wind power plant cluster, and calculating the neighborhood coupling factor of the wind power plant cluster; and constructing a wind power plant parameter distribution model, and formulating a corresponding current transmission distribution strategy for each wind power plant according to the marine environment complexity index and the neighborhood coupling factor of each wind power plant. According to the invention, the problem that the real-time performance of a control system is seriously influenced by communication delay and data loss in an open-sea wind power plant is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore power transmission control, and particularly to an optimized control method for flexible DC power transmission applicable to the access of offshore wind power. Background Art

[0002] Since offshore wind farms are often far from the shore and have complex environments, the access of offshore wind farms to onshore power grids and large-scale long-distance power transmission face a series of challenges. As an advanced power transmission technology for solving long-distance and multi-terminal interconnection problems, flexible DC power transmission has become an important technical means for the access of offshore wind power by virtue of its flexible regulation, commutation without failure, and independent control of active and reactive power. However, in practical applications, there are still some deficiencies and technical problems in this technology: Flexible DC power transmission systems usually involve the coordinated control of offshore converter stations and onshore converter stations, as well as multiple units within the wind farm. High-precision and high-speed information interaction and coordination are required between control units to ensure the stable operation of the entire system under fault and fluctuating conditions. Existing methods still have problems such as information delay and inconsistent control strategies in coordinating the AC side and the DC side, as well as the coordinated scheduling between control units at all levels, which will directly affect the power distribution and stable regulation of the system.

[0003] There are natural uncertainties in offshore wind farms. For example, environmental parameters such as wind speed, sea current, and wave have high randomness and time-variability. Existing control methods often rely on offline modeling or predictive control based on fixed parameters, and it is difficult to capture real-time environmental changes in a timely manner, resulting in untimely responses of power transmission control strategies in the face of emergencies and even causing system instability. In addition, data acquisition and communication delay are also problems that need to be urgently solved in current practical engineering.

[0004] Therefore, the present invention provides an optimized control method for flexible DC power transmission applicable to the access of offshore wind power. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimized control method and system for flexible DC power transmission applicable to the access of offshore wind power to solve the existing problems mentioned in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solution: An optimized control method for flexible DC power transmission applicable to the access of offshore wind power, comprising the following steps: S1. Collect the longitude and latitude coordinates and the distance from the shore of each wind farm, and collect the environmental parameters of each wind farm; S2. Set a prediction strategy for the complexity of the sea surface environment to obtain an index of the complexity of the marine environment; S3. Calculate the relative positions and mutual dependencies of each wind farm in the entire region according to the spatial distribution of the wind farms to obtain the position of the wind farm cluster, and calculate its neighborhood coupling factor; S4. Construct a wind farm parameter allocation model, and formulate corresponding current transmission allocation strategies for each wind farm according to the marine environment complexity index and neighborhood coupling factor of each wind farm.

[0007] A further improvement of the present invention lies in that the sea surface environment complexity prediction strategy includes a historical complexity coefficient construction sub-strategy, a perturbation factor construction sub-strategy, an underwater acoustic resonance index sub-strategy, and a marine environment complexity index calculation sub-strategy; the historical complexity coefficient construction sub-strategy includes collecting the energy utilization rate during the historical operation of each wind farm, and defining the historical complexity coefficient HCC through the standardized offshore distance and the standardized energy utilization rate; the perturbation factor construction sub-strategy includes recording the sea surface environment data, where the sea surface environment data includes the sea surface wave height, wind speed, seawater temperature, and salt fog concentration of each wind farm, and mapping the sea surface environment data into a non-linear function to obtain the perturbation factor ref.

[0008] A further improvement of the present invention lies in that the underwater acoustic resonance index sub-strategy includes obtaining the sea wave excitation frequency band and the ocean current excitation frequency band ; the specific process includes: collecting the historical sea wave period, and obtaining the sea wave excitation frequency band by calculating the reciprocal of the mean of the collected historical sea wave periods; deploying an underwater acoustic sensor array to collect the ocean ambient noise signal, and the specific process includes performing a fast Fourier transform on the ocean ambient noise signal to obtain the ocean sound spectrum S(f), dividing the spectrum into N frequency bands according to a set step size, calculating the correlation coefficient between all frequency bands and the perturbation factor, and extracting the frequency bands with the correlation coefficient greater than the set correlation coefficient threshold as the ocean current excitation frequency band; combining the sea wave excitation frequency band and the ocean current excitation frequency band to obtain the acoustic-electric coupling term Ari, and the calculation formula is: where, represents the i-th ocean current excitation frequency band, and n represents the number of ocean current excitation frequency bands.

[0009] A further improvement of the present invention lies in that the marine environment complexity index calculation sub-strategy constructs a marine environment complexity index by combining the historical complexity coefficient, the perturbation factor, and the acoustic-electric coupling term, and is specifically expressed as: ; where, represents the normalized acoustic-electric coupling term, represents the set acoustic-electric coupling gain, represents the non-linear mapping function.

[0010] A further improvement of the present invention lies in that the specific steps of step S3 include: S31. Take the longitude and latitude coordinates of each wind farm as input vectors and send them into the wind farm zoning strategy to obtain M wind farm clusters; S32. For each wind farm cluster β, calculate its neighborhood coupling factor , where is the Euclidean distance between wind farm α and β, and α represents any wind farm in wind farm cluster β, represents the set characteristic scale parameter.

[0011] A further improvement of the present invention lies in that the wind farm zoning strategy includes a preliminary mapping output layer and an abstract clustering output layer; the specific steps include: S311. The preliminary mapping output layer maps the original spatial coordinates to a two-dimensional space, and uses competitive learning to form K candidate wind farm clusters for adjacent wind farms; S312. The abstract clustering output layer extracts the centers of the K candidate wind farm clusters ; S313. For each wind farm, find the center of the candidate wind farm cluster closest to it through the Euclidean distance , and let participate in competitive update; S314. Coordinately update the original cluster center of each wind farm and through the neighborhood function; S315. When the absolute value of the difference between the marine environment complexity index of the updated wind farm cluster center and the mean value of the marine environment complexity index of the wind farm cluster is less than the set cluster difference threshold, stop the competition.

[0012] A further improvement of the present invention lies in that the wind farm parameter allocation model includes defining the current transmission allocation weight of each wind farm cluster , where represents the mean value of the marine environment complexity index of the β-th wind farm cluster; extract the next transmission-back cluster set of each wind farm cluster. At this time, calculate the current transmission allocation weight of each wind farm cluster in the next transmission-back cluster set, and allocate the transmission current of each wind farm cluster according to the current transmission allocation weight of each wind farm cluster in the next transmission-back cluster set.

[0013] A further improvement of the present invention lies in that the wind farm parameter allocation model further includes an adaptive marine environment complexity index update sub-model; the adaptive marine environment complexity index update sub-model extracts the mean value of the marine environment complexity index and the mean value of the sound-electric coupling term of the β-th wind farm cluster in the past year, and sets a target gain function F, expressed as , where represents the scaling factor, represents the environmental benchmark threshold. When =MECI, output the historical MECI intermediate value of the β-th wind farm cluster, represents the Ari gain weight, represents the gain function. Integrate the target gain function into the wind farm controller and perform adaptive adjustment on through discrete time.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: First, the present invention realizes acoustic-electric coupling control by introducing underwater acoustic feedback, can identify the early warning signals of drastic environmental changes in advance, and adjusts the current transmission strategy of the wind farm through adaptive control, thereby reducing the risks under extreme working conditions and protecting the equipment from failure due to overload or abnormal environmental impacts; Introduce the neighborhood coupling factor between wind farms, and use the distributed cooperative optimization algorithm to calculate the dynamic weights of each node in real time, so that the current distribution strategy can fully reflect the relative positions and cluster densities of each wind farm in the entire network, which helps to reduce the local operation risks while balancing the transmission load in the whole region and improving the overall stability and economy of the system; By deploying edge computing units at each wind farm node, data preprocessing, index calculation and local decision-making are all completed on-site, and overall coordination is achieved through distributed algorithms, thereby greatly improving the real-time performance and fault tolerance ability, and ensuring that the system can still operate stably when communication is abnormal or node failures occur. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flowchart of an optimized control method for flexible DC transmission applicable to offshore wind power access according to the present invention; Figure 2 is a flowchart of calculating the neighborhood coupling factor of an optimized control method for flexible DC transmission applicable to offshore wind power access according to the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The technical solution of the present invention will be described in detail below with reference to the drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. Without conflict, the technical features in the embodiments of the present invention and the embodiments can be combined with each other.

[0017] The term "and / or" merely describes the associated relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects before and after.

[0018] Embodiment 1 Figure 1The figure shows a flowchart of an optimized control method for flexible DC transmission applicable to offshore wind power access disclosed in this embodiment. The steps are as follows: S1. Collect the longitude and latitude coordinates and the offshore distance of each wind farm, and collect the environmental parameters of each wind farm; S2. Set a prediction strategy for the complexity of the sea surface environment to obtain an index of the complexity of the marine environment; S3. Calculate the relative position and mutual dependence of each wind farm in the entire area according to the spatial distribution of the wind farms to obtain the location of the wind farm cluster, and calculate its neighborhood coupling factor; S4. Construct a parameter allocation model for the wind farms, and formulate corresponding current transmission allocation strategies for each wind farm according to the marine environment complexity index and the neighborhood coupling factor of each wind farm.

[0019] The prediction strategy for the complexity of the sea surface environment includes a historical complexity coefficient construction sub-strategy, a perturbation factor construction sub-strategy, an underwater acoustic resonance index sub-strategy, and a marine environment complexity index calculation sub-strategy; The historical complexity coefficient construction sub-strategy includes collecting the energy utilization rate during the historical operation of each wind farm, and defining the historical complexity coefficient HCC through the standardized offshore distance and the standardized energy utilization rate; , where represents the value of the standardized offshore distance, d represents the offshore distance, represents the value of the standardized mean of the historical energy utilization rate, represents the mean of the historical energy utilization rate, represents a regulation parameter used to amplify the non-linear effect of the ratio of the two; The perturbation factor construction sub-strategy includes recording the sea surface wave height Hsu, wind speed Usw, seawater temperature and salt spray concentration Ssc of each wind farm, and mapping the sea surface environment data into a non-linear function to obtain the perturbation factor ref; specifically expressed as: ; where a, b, c, d, e are design constants, and the non-linear and periodic characteristics of the sea conditions are captured through non-linear functions such as sin() and exp().

[0020] The underwater acoustic resonance index sub-strategy includes obtaining the sea wave excitation frequency band and the sea current excitation frequency band ; The specific process includes: Collect the historical sea wave periods, and obtain the sea wave excitation frequency band by calculating the reciprocal of the mean of the collected historical sea wave periods; affected by wind, tide and waves, the perturbation of the sea waves is usually concentrated in the medium and low frequency bands; Deploy an underwater acoustic sensor array to collect ocean ambient signals. The specific process includes performing a fast Fourier transform on the ocean ambient signals to obtain the ocean sound spectrum S(f), dividing the spectrum into N frequency bands at a set step size. Since ocean currents are affected by parameters such as wave height, wind speed, seawater temperature, and salt spray concentration, the ocean current excitation frequency bands are obtained by calculating the correlation coefficients between all frequency bands of the spectrum and the wind speed, calculating the correlation coefficients between all frequency bands and the perturbation factors, and extracting the frequency bands with correlation coefficients greater than the set correlation coefficient threshold as the ocean current excitation frequency bands; Combine the ocean wave excitation frequency bands and the ocean current excitation frequency bands to obtain the acoustic-electric coupling term Ari. The calculation formula is: , where represents the i-th ocean current excitation frequency band, and n represents the number of ocean current excitation frequency bands.

[0021] The calculation sub-strategy of the ocean environment complexity index constructs the ocean environment complexity index by combining the historical complexity coefficient, the perturbation factor, and the acoustic-electric coupling term. Previous systems often only considered a part of the indicators in real-time data or historical data in the transmission current allocation, while ignoring the coupling effect between the two. The present invention first constructs the historical complexity coefficient HCC generated by the offshore distance and the power generation utilization efficiency using historical data, and then combines the real-time environmental perturbation ref and the acoustic feedback, organically coupling the two to form a comprehensive ocean environment complexity index, which is specifically expressed as: ; where represents the normalized acoustic-electric coupling term, represents the set acoustic-electric coupling gain, represents the non-linear mapping function, expressed as , which is used to convert the acoustic feedback into an adjustment term, represents the set reference value of the acoustic-electric coupling term.

[0022] Embodiment 2 Based on the technical solution of Embodiment 1, this embodiment provides the specific calculation process of the neighborhood coupling factor in step S3. Figure 2 shows the calculation flow chart of the neighborhood coupling factor of a flexible DC transmission optimization control method applicable to offshore wind power access disclosed in this embodiment. The steps are as follows: S31. Take the longitude and latitude coordinates of each wind farm as input vectors and send them into the wind farm zoning strategy to obtain M wind farm clusters, so that the wind farms within each cluster have a relatively close geographical distance and high similarity; The wind farm zoning strategy includes a preliminary mapping output layer and an abstract clustering output layer; the specific steps include: S311. The preliminary mapping output layer maps the original spatial coordinates to a two-dimensional space, and uses competitive learning to form K candidate wind farm clusters for adjacent wind farms. S312. The abstract clustering output layer extracts the centers of the K candidate wind farm clusters ; S313. For each wind farm, find the center of the candidate wind farm cluster closest to it through the Euclidean distance , and let participate in the competitive update, and the update formula is: ; where represents the set learning rate that gradually decays with iteration. Only when the current input wind farm is closest to the center of a certain wind farm cluster candidate , will this center be selected as the "winner" and updated. S314. To enable adjacent candidate centers to be activated simultaneously, the original cluster center of each wind farm and are co-updated through the neighborhood function: ; where, represents the neighborhood function; the neighborhood function in the present invention is a Gaussian function.

[0023] S315. When the absolute value of the difference between the marine environment complexity index of the updated wind farm cluster center and the mean value of the marine environment complexity index of this wind farm cluster is less than the set cluster difference threshold, stop the competition.

[0024] The wind farm zoning strategy has an adaptive learning ability, can dynamically adjust the wind farm cluster clustering results according to real-time data, so that stable clustering centers can be obtained even when the position data noise of the offshore wind farm is large, effectively reducing the system regulation error; Using the wind farm cluster clustering results, a collaborative mechanism for different nodes within the region can be constructed to achieve distributed optimization, ensuring that when the risk in the local dense area is high, the transmission current in this area is automatically reduced, while fully exerting the output in the local sparse area to achieve global optimal allocation.

[0025] S32. For each wind farm cluster, name it with its cluster center β, and calculate its neighborhood coupling factor , where, is the Euclidean distance between wind farm α and β, α represents any wind farm in wind farm cluster β, represents the set characteristic scale parameter, which is used to control the attenuation rate of the distance influence. The neighborhood coupling factor represents the superposition of exponential decay functions, and the closer nodes contribute more to the factor, while the contribution of distant nodes decays; In this embodiment, the spatial relationship between each wind farm is comprehensively considered in a non-linear manner to extract local density information, and each wind farm cluster can calculate through local information and communication with neighboring nodes. This reduces the pressure of centralized computing; moreover, the neighborhood coupling factor can reflect the degree of interdependence within the local cluster, providing a basis for introducing cooperative control into the current distribution strategy.

[0026] Embodiment 3 Based on the technical solutions of Embodiment 1 and Embodiment 2, this embodiment provides the specific construction process of the wind farm parameter allocation model in step S4, as follows: The wind farm parameter allocation model includes defining the current transmission allocation weight for each wind farm cluster. Among them, represents the mean value of the marine environment complexity index of the β-th wind farm cluster; according to the current transmission allocation weight of the wind farm clusters through which the current return path of each wind farm cluster passes, the transmission current is allocated. For example, the A wind farm cluster can select Q wind farm clusters as the next return passing points. At this time, calculate the current transmission allocation weights of these Q wind farm clusters, calculate the proportion of the current transmission allocation weight of the λ-th sub-power plant cluster in the Q wind farm clusters in the total current transmission allocation weight of the Q wind farm clusters, and multiply it by the total current transmission of the A wind farm cluster to obtain the transmission current of this wind farm cluster.

[0027] When the MECI is high, that is, the environmental complexity risk is large, the weight decreases; when the neighborhood coupling factor is high, it indicates that this wind farm cluster is located in a relatively dense area and there is a cooperative complementary effect. Therefore, the single-point weight is reduced to achieve global cooperative allocation.

[0028] In view of the situation that each wind farm cluster may not be able to obtain global information in extreme environments or communication loss, the wind farm parameter allocation model is equipped with an adaptive marine environment complexity index update sub-model. The adaptive marine environment complexity index update sub-model extracts the mean value of the marine environment complexity index and the mean value of the acoustic-electric coupling term of the β-th wind farm cluster in the past year, and sets a target gain function F, expressed as Among them, represents the proportionality factor, which is used to adjust the influence of MECI on the output. represents the environmental benchmark threshold. In this way, the function F will automatically map the current environment and acoustic state to a desired coupling gain target value. When =MECI, the intermediate value of the historical MECI of the β-th wind farm cluster is output. represents the Ari gain weight, which is used to amplify the sensitivity of Ari changes. Represents the gain function, denoted as so as to cause violent fluctuations when Ari is relatively large; Integrate the target gain function into the wind farm controller, and perform adaptive adjustment on through discrete time; The adjustment process is expressed as: wherein, represents the learning rate, which determines the update amplitude. Selecting an appropriate can quickly respond to local environmental changes while maintaining system stability. The update formula adopts the idea similar to discrete-time first-order filtering for smooth non-linear adjustment.

[0029] In this embodiment, a distributed collaborative network is constructed among wind farm clusters. Even in the case of communication interruption, each node can independently run the adaptive ocean environment complexity index update sub-model, and automatically fuse the information of each wind farm cluster when communication is restored to maintain global consistency.

[0030] The set values such as the threshold and weight can be set according to the default settings of the present invention, or can be set by the operator himself.

[0031] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented 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.

[0032] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0033] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions in the processFigure 1 one process or multiple processes and / or boxes Figure 1 the functions specified in one box or multiple boxes

[0034] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or boxes Figure 1 one box or multiple boxes

[0035] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the claims of the present invention. These all fall within the protection scope of the present invention.

Claims

1. A flexible direct current transmission optimization control method suitable for offshore wind power access, characterized in that: The following steps are involved: S1. Collect the longitude and latitude coordinates and offshore distance of each wind farm, and collect the environmental parameters of each wind farm; S2. Setting a sea surface environment complexity prediction strategy to obtain a marine environment complexity index; S3. Calculate the relative position and interdependence of each wind farm in the entire region according to the spatial distribution of wind farms, obtain the location of the wind farm cluster, and calculate its neighborhood coupling factor; S4. Construct a wind farm parameter allocation model and formulate a corresponding current transmission allocation strategy for each wind farm based on the marine environment complexity index and neighborhood coupling factor of each wind farm.

2. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 1, characterized in that: The sea surface environment complexity prediction strategy includes a historical complexity coefficient construction sub-strategy, a disturbance factor construction sub-strategy, an underwater acoustic resonance index sub-strategy and a marine environment complexity index calculation sub-strategy; the historical complexity coefficient construction sub-strategy includes collecting the energy utilization rate of each wind farm during the historical operation period, and defining the historical complexity coefficient HCC through the standardized offshore distance and the standardized energy utilization rate; The disturbance factor construction sub-strategy includes recording sea surface environment data, which includes sea surface wave height, wind speed, sea water temperature and salt spray concentration of each wind farm, and mapping the sea surface environment data into a nonlinear function to obtain a disturbance factor ref.

3. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 2 is characterized in that: The underwater acoustic resonance index sub-strategy includes obtaining the wave excitation frequency band and ocean current excitation frequency band ; The specific process includes: collecting historical wave cycles, and obtaining the wave excitation frequency band by calculating the inverse of the mean of the collected historical wave cycles; deploying an underwater acoustic sensor array to collect ocean wake signals, and the specific process includes performing fast Fourier transform on the ocean wake signals to obtain the ocean sound spectrum S(f), dividing the ocean sound spectrum into N frequency bands according to the set step size, calculating the correlation coefficients of all frequency bands and disturbance factors, and extracting the frequency bands with correlation coefficients greater than the set correlation coefficient threshold as the current excitation frequency bands; combining the wave excitation frequency bands and the current excitation frequency bands to obtain the acoustic-electric coupling term Ari, and the calculation formula is: ,in, represents the i-th ocean current excitation frequency band, and n represents the number of ocean current excitation frequency bands.

4. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 2 is characterized in that: The marine environment complexity index calculation sub-strategy constructs the marine environment complexity index by combining the historical complexity coefficient, the disturbance factor and the acoustic-electric coupling term, which is specifically expressed as: ； in, represents the normalized acoustic-electric coupling term, Indicates the set acoustic-electric coupling gain, Represents a nonlinear mapping function.

5. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 1 is characterized in that: Step S3 specifically includes: S31, taking the longitude and latitude coordinates of each wind farm as an input vector and sending it into the wind farm zoning strategy to obtain M wind farm clusters; S32. For each wind farm cluster β, calculate its neighborhood coupling factor ,in, is the Euclidean distance between wind farms α and β, α represents any wind farm in wind farm cluster β, Represents the set characteristic scale parameter.

6. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 5 is characterized in that: The wind farm zoning strategy includes a preliminary mapping output layer and an abstract clustering output layer; The specific steps include: S311, the preliminary mapping output layer maps the original space coordinates to a two-dimensional space, and uses competitive learning to make adjacent wind farms form K candidate wind farm clusters; S312, the abstract clustering output layer extracts K candidate wind farm cluster centers ; S313. For each wind farm, find the candidate wind farm cluster center closest to it by Euclidean distance. , and order Participate in competition updates; S314, the original cluster center and Conduct collaborative updates; S315. When the absolute value of the difference between the updated marine environment complexity index of the wind farm cluster center and the mean marine environment complexity index of the wind farm cluster is less than the set cluster difference threshold, stop the competition.

7. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 5 is characterized in that: The wind farm parameter allocation model includes defining the current transmission allocation weights of each wind farm cluster ,in Represents the mean value of the marine environment complexity index of the βth wind farm cluster; extracts the next transmission cluster set of each wind farm cluster, at this time, calculates the current transmission allocation weight of each wind farm cluster in the next transmission cluster set, and allocates the transmission current of each wind farm cluster according to the current transmission allocation weight of each wind farm cluster in the next transmission cluster set.

8. The method for optimizing and controlling flexible direct current transmission applicable to offshore wind power access according to claim 7 is characterized in that: The wind farm parameter allocation model also includes an adaptive marine environment complexity index update sub-model; the adaptive marine environment complexity index update sub-model extracts the mean value of the marine environment complexity index and the mean value of the acoustic-electric coupling term of the β-th wind farm cluster in the past year, and sets a target gain function F, which is expressed as ,in, represents the scale factor, Indicates the environmental benchmark threshold. = MECI, output the historical MECI median value of the β-th wind farm cluster, represents the Ari gain weight, Represents the gain function, integrates the target gain function into the wind farm controller, and uses discrete time Make adaptive adjustments.

Images