Optimization method and device for mooring system of wind power plant, equipment and medium

Abstract

The invention provides a wind power plant mooring system optimization method and device, equipment and a medium. The method comprises the steps that a sea condition classification model is constructed based on obtained historical sea condition information of a wind power plant; based on a sea condition category output by the sea condition classification model, obtaining a corresponding mooring rope parameter; constructing a sea condition-mooring rope parameter database based on the sea condition categories and the mooring rope parameters, wherein the database at least comprises a corresponding relation between the sea condition categories and the mooring rope parameters; inputting the obtained current sea condition information into a sea condition classification model to obtain a current sea condition category corresponding to the current sea condition information; and obtaining target mooring line parameters corresponding to the current sea condition category based on a sea condition-mooring line parameter database. According to the method, the sea condition classification model is constructed by using the historical sea condition information, and the mooring cable parameters can be dynamically adjusted according to the real-time sea condition information in combination with the sea condition-mooring cable parameter database containing the corresponding relationship between the sea condition categories and the mooring cable parameters, so that the mooring requirements under different sea conditions are accurately matched.

Description

Technical Field

[0001] This disclosure relates to the field of wind farm technology, and in particular to an optimization method, apparatus, equipment and medium for wind farm mooring systems. Background Technology

[0002] Existing mooring systems are mostly designed statically based on specific sea conditions, making it difficult to adapt to the complex and ever-changing marine environment. Loads such as wind, waves, and currents in the marine environment are constantly changing, and marine environmental parameters vary significantly across different seasons and times. The mooring cable length in traditional mooring systems is usually fixed and cannot be dynamically adjusted based on real-time marine environmental data. This can lead to excessive tension on the mooring cable under extreme sea conditions, increasing the risk of breakage. Conversely, when sea conditions are relatively stable, the mooring cable may be too long, causing excessive displacement of the wind turbine and affecting power generation efficiency. This is because traditional design methods do not fully consider the dynamic characteristics of the marine environment and lack in-depth mining and utilization of historical marine environmental data, making it difficult to establish a mooring system model that can respond to environmental changes in real time. Summary of the Invention

[0003] The technical problem to be solved by this disclosure is that the mooring cables in the prior art are usually of fixed length and cannot be dynamically adjusted according to real-time marine environmental data. The disclosure provides an optimization method, device, equipment and medium for wind farm mooring systems.

[0004] This disclosure solves the above-mentioned technical problems through the following technical solution:

[0005] The first aspect of this disclosure provides an optimization method for a wind farm mooring system, the optimization method comprising:

[0006] Obtain historical sea condition information for wind farms;

[0007] A sea state classification model is constructed based on the historical sea state information;

[0008] Based on the sea state category output by the sea state classification model, obtain the corresponding mooring cable parameters;

[0009] A sea state-mooring cable parameter database is constructed based on the sea state categories and mooring cable parameters. The sea state-mooring cable parameter database contains at least the correspondence between sea state categories and mooring cable parameters.

[0010] Obtain current sea state information for the wind farm;

[0011] The current sea state information is input into the sea state classification model to obtain the current sea state category corresponding to the current sea state information;

[0012] Based on the sea state-mooring cable parameter database, obtain the target mooring cable parameters corresponding to the current sea state category.

[0013] Preferably, the optimization method further includes:

[0014] Obtain the current mooring cable parameters corresponding to the current sea state category;

[0015] In response to the inconsistency between the current mooring cable parameters and the target mooring cable parameters, the current mooring cable parameters are adjusted based on the target mooring cable parameters to make the current mooring cable parameters consistent with the target mooring cable parameters.

[0016] Preferably, the mooring cable parameters include mooring cable length and mooring damping, and adjusting the current mooring cable parameters based on the target mooring cable parameters to make the current mooring cable parameters consistent with the target mooring cable parameters includes:

[0017] The current mooring cable length and the current mooring damping are adjusted by the mooring adjustment unit to make the current mooring cable length consistent with the target mooring cable length and the current mooring damping consistent with the target mooring damping.

[0018] Preferably, the step of constructing a sea state classification model based on the historical sea state information includes:

[0019] Machine learning and deep learning algorithms suitable for time series classification were selected as training algorithms;

[0020] The historical sea state information is input into the training algorithm to obtain the sea state classification model.

[0021] Preferably, inputting the historical sea state information into the training algorithm to obtain the sea state classification model includes the following steps:

[0022] The historical sea state information is divided according to a preset ratio to obtain a training set, a validation set, and a test set;

[0023] The training set is input into the training algorithm to obtain the sea state classification model;

[0024] The validation set is input into the sea state classification model to adjust the parameters of the sea state classification model based on the validation results;

[0025] The test set is input into the sea state classification model to test the sea state classification results of the sea state classification model, so as to optimize the sea state classification model.

[0026] Preferably, the step of inputting the test set into the sea state classification model and testing the sea state classification results of the sea state classification model to optimize the sea state classification model includes:

[0027] Obtain the actual sea state category and the target sea state category corresponding to the test set;

[0028] A loss function is obtained based on the target sea state category and the actual sea state category;

[0029] The sea state classification model is optimized based on the loss function.

[0030] Preferably, the step of constructing a sea state classification model based on the historical sea state information further includes:

[0031] The historical sea state information is processed by at least one of the following methods: timestamp alignment, spatial interpolation, coordinate transformation, outlier detection, and missing value imputation, to obtain preprocessed historical sea state information.

[0032] Preferably, the historical sea condition information includes at least one of historical wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current speed information, and ocean current direction information.

[0033] Preferably, the optimization method further includes:

[0034] The sea state classification model and the sea state-mooring cable parameter database are stored in the cloud network or on the wind turbine side.

[0035] Preferably, obtaining the current sea state information of the wind farm includes:

[0036] Based on the data acquisition equipment on the wind farm or wind turbine side, at least one of the following information is collected: current wind speed, wind direction, wave height, wave period, wave direction, ocean current velocity, and ocean current direction. The data acquisition equipment includes at least one of the following: a wind measurement tower, an acoustic Doppler current meter, and a laser wave meter.

[0037] A second aspect of this disclosure provides an optimization device for a wind farm mooring system, the optimization device comprising:

[0038] The first acquisition module is used to acquire historical sea condition information of the wind farm;

[0039] The first construction module is used to construct a sea state classification model based on the historical sea state information;

[0040] The second acquisition module is used to acquire the corresponding mooring cable parameters based on the sea state category output by the sea state classification model.

[0041] The second construction module is used to construct a sea state-mooring cable parameter database based on the sea state category and mooring cable parameters. The sea state-mooring cable parameter database contains at least the correspondence between sea state categories and mooring cable parameters.

[0042] The third acquisition module is used to acquire the current sea state information of the wind farm;

[0043] The fourth acquisition module is used to input the current sea state information into the sea state classification model to obtain the current sea state category corresponding to the current sea state information;

[0044] The fifth acquisition module is used to acquire the target mooring cable parameters corresponding to the current sea state category based on the sea state-mooring cable parameter database.

[0045] Preferably, the optimization device further includes:

[0046] The sixth acquisition module is used to acquire the current mooring cable parameters corresponding to the current sea state category;

[0047] An adjustment module is configured to adjust the current mooring cable parameters based on the target mooring cable parameters in response to a discrepancy between the current mooring cable parameters and the target mooring cable parameters, so that the current mooring cable parameters are consistent with the target mooring cable parameters.

[0048] Preferably, the mooring cable parameters include mooring cable length and mooring damping. The adjustment module is used to adjust the current mooring cable length and the current mooring damping through the mooring adjustment unit so that the current mooring cable length is consistent with the target mooring cable length and the current mooring damping is consistent with the target mooring damping.

[0049] Preferably, the first building module includes:

[0050] The selection unit is used to select machine learning and deep learning algorithms that can be used for time series classification as training algorithms;

[0051] The acquisition unit is used to input the historical sea state information into the training algorithm to obtain the sea state classification model.

[0052] Preferably, the acquisition unit includes:

[0053] The historical sea state information is divided into sub-units according to a preset ratio to obtain a training set, a validation set, and a test set.

[0054] Obtain sub-units for inputting the training set into the training algorithm to obtain the sea state classification model;

[0055] An adjustment subunit is used to input the validation set into the sea state classification model in order to adjust the parameters of the sea state classification model based on the validation results;

[0056] The testing unit is used to input the test set into the sea state classification model, test the sea state classification results of the sea state classification model, and optimize the sea state classification model.

[0057] Preferably, the testing unit is used to obtain the real sea state category and the target sea state category corresponding to the test set; obtain a loss function based on the target sea state category and the real sea state category; and optimize the sea state classification model based on the loss function.

[0058] Preferably, the first building module further includes:

[0059] The preprocessing unit is used to perform at least one of the following on the historical sea state information: timestamp alignment, spatial interpolation, coordinate transformation, outlier detection, and missing value filling, to obtain preprocessed historical sea state information.

[0060] Preferably, the historical sea condition information includes at least one of historical wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current speed information, and ocean current direction information.

[0061] Preferably, the optimization device further includes:

[0062] The storage module is used to store the sea state classification model and the sea state-mooring cable parameter database to the cloud network or the wind turbine side.

[0063] Preferably, the third acquisition module is used to acquire at least one of the following information based on the data acquisition equipment on the wind farm or wind turbine side: current wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current velocity information, and ocean current direction information of the wind farm. The data acquisition equipment includes at least one of the following: a wind measurement tower, an acoustic Doppler current meter, and a laser wave meter.

[0064] A third aspect of this disclosure provides an electronic device including a memory, a processor, and a computer program stored in the memory and for running on the processor, wherein the processor executes the computer program to implement the optimization method for the wind farm mooring system described in the first aspect.

[0065] The fourth aspect of this disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the optimization method for the wind farm mooring system described in the first aspect.

[0066] The fifth aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the optimization method for a wind farm mooring system as described in the first aspect.

[0067] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this disclosure.

[0068] The positive and progressive effects of this disclosure are as follows:

[0069] This disclosure utilizes historical sea state information to construct a sea state classification model, and combines it with a sea state-mooring cable parameter database containing the correspondence between sea state categories and mooring cable parameters. This allows for dynamic adjustment of mooring cable parameters based on real-time sea state information, thereby accurately matching mooring requirements under different sea states. Attached Figure Description

[0070] Figure 1 A flowchart of an optimization method for a wind farm mooring system provided in Embodiment 1 of this disclosure.

[0071] Figure 2 This is a schematic diagram of the module of the optimization device for the wind farm mooring system provided in Embodiment 2 of this disclosure.

[0072] Figure 3 This is a schematic diagram of the electronic device used to implement the optimization method for a wind farm mooring system according to Embodiment 3 of this disclosure. Detailed Implementation

[0073] The present disclosure is further illustrated below by way of embodiments, but the present disclosure is not limited to the scope of the embodiments described herein.

[0074] The prefixes such as "first" and "second" used in this disclosure are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this disclosure does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not be construed as an unnecessary limitation. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0075] In this embodiment of the disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good morals.

[0076] Example 1

[0077] Figure 1 A flowchart of an optimization method for a wind farm mooring system provided in Embodiment 1 of this disclosure is shown below. Figure 1 As shown, the optimization method includes:

[0078] S1. Obtain historical sea condition information for the wind farm;

[0079] In this embodiment, based on the location of the wind farm, historical sea state information for the sea area where the wind farm is located over the past several years (e.g., the past 10 years) is downloaded from NOAA (National Oceanic and Atmospheric Administration).

[0080] In one optional implementation, the historical sea state information includes at least one of historical wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current speed information, and ocean current direction information.

[0081] S2. Construct a sea state classification model based on historical sea state information;

[0082] In this embodiment, the network structure of the sea state classification model can be a CNN-LSTM network. The input layer of the CNN-LSTM network is a 7-dimensional temporal vector (time window = 60 min, stride = 15 min), the CNN (Convolutional Neural Network) convolution kernel size is 5*1, the LSTM (Long Short-Term Memory) unit is 256, the network has 3 fully connected layers (Dropout = 0.3), and the activation function is softmax = 4.

[0083] S3. Based on the sea state category output by the sea state classification model, obtain the corresponding mooring cable parameters;

[0084] S4. Construct a sea state-mooring cable parameter database based on sea state categories and mooring cable parameters. The sea state-mooring cable parameter database shall at least contain the correspondence between sea state categories and mooring cable parameters.

[0085] In this embodiment, the mooring cable parameters include the mooring cable length and mooring damping;

[0086] In a specific example, taking mooring cable length as an example, the most severe operating parameters for each sea state category—wind, waves, and current—are selected to prepare for the design of the mooring cable length. For instance, the sea state classification model outputs four sea state categories (C1, C2, C3, and C4). Numerical simulation software (such as SIMA and AQWA) is used to design the corresponding mooring cable lengths. Following standard procedures, the mooring cable lengths corresponding to each sea state category are calculated based on water depth and the floating platform. Specifically, a set of "preset tension" and "delivery length" for a set of mooring cables are initially assumed based on sea state, platform, water depth, and mooring cable properties. Under the action of "average environmental load," the floating platform will shift, and the floating platform's displacement is calculated. When the platform is at a certain offset position, the tension provided by all mooring cables is used to further determine the equilibrium position of the floating platform, ensuring that the sum of the tensions of all mooring cables is equal in magnitude and opposite in direction to the average environmental load, and that the torques are balanced. Then, the average offset of the floating platform under this sea state category, the top tension, angle, and bottom tension of each mooring cable, as well as the seabed morphology of the mooring cables are output. The average offset of the floating platform under this sea state category, the top tension, angle, and bottom tension of each mooring cable, and the seabed morphology of the mooring cables are checked to see if they all meet the specifications and safety requirements. If at least one is not met, the "preset tension" or "cable length" is adjusted. If it is met, the mooring cable length corresponding to this sea state category is output.

[0087] As shown in Table 1, calculate the mooring cable lengths L1, L2, L3, and L4 corresponding to the four sea state categories C1, C2, C3, and C4, respectively. It should be noted that the mooring cable length is the length that minimizes the motion response of the floating platform and provides sufficient restoring force for mooring.

[0088] Table 1

[0089]

[0090] As shown in Table 1, each sea state category has a corresponding mooring cable length, meaning there is a correspondence between sea state categories and mooring cable lengths. Furthermore, the correspondence between sea state categories and mooring cable parameters is stored in the sea state-mooring cable parameter database.

[0091] S5. Obtain the current sea state information of the wind farm;

[0092] In an optional implementation, the current sea state information of the wind farm is obtained by means of the following method: specifically, based on the data acquisition equipment on the wind farm or wind turbine side, at least one of the following is collected: current wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current velocity information, and ocean current direction information of the wind farm. The data acquisition equipment includes at least one of the following: a wind measurement tower, an acoustic Doppler current meter, and a laser wave meter.

[0093] In this embodiment, wind speed and wind direction information of the wind farm are measured by a wind measuring tower, ocean current velocity and ocean current direction information are measured by an acoustic Doppler current meter, and wave height, wave period and wave direction information are measured by a laser wave meter.

[0094] S6. Input the current sea state information into the sea state classification model to obtain the current sea state category corresponding to the current sea state information;

[0095] S7. Obtain the target mooring cable parameters corresponding to the current sea state category based on the sea state-mooring cable parameter database.

[0096] In this embodiment, the current sea state information is input into the sea state classification model to obtain the current sea state category corresponding to the current sea state information; further, the target mooring cable parameters corresponding to the current sea state information are obtained based on the sea state-mooring cable parameter database.

[0097] In an optional implementation, the optimization method further includes:

[0098] S31. Obtain the current mooring cable parameters corresponding to the current sea state category;

[0099] S32. In response to the inconsistency between the current mooring cable parameters and the target mooring cable parameters, adjust the current mooring cable parameters based on the target mooring cable parameters to make the current mooring cable parameters consistent with the target mooring cable parameters.

[0100] In an optional implementation, the mooring cable parameters include mooring cable length and mooring damping. S32 includes: adjusting the current mooring cable length and current mooring damping through the mooring adjustment unit so that the current mooring cable length is consistent with the target mooring cable length and the current mooring damping is consistent with the target mooring damping.

[0101] In this embodiment, when the current mooring cable length is inconsistent with the target mooring cable length and / or the current mooring damping is inconsistent with the target mooring damping, the wind farm mooring system activates the mooring adjustment unit to adjust the current mooring cable length and the current mooring damping so that the current mooring cable length and the current mooring damping are consistent with the target mooring cable length and the target mooring damping.

[0102] It should be noted that in this embodiment, the length of the mooring cable and the mooring damping can be adjusted by the mooring adjustment unit (for example, the mooring adjustment unit can be a conventional tensioner, anchor winch, chain stopper, etc.). Alternatively, the wind farm mooring system can be equipped with an active damping control unit to adjust the mooring damping according to different sea conditions and the length of the mooring cable. For example, in the case of severe sea conditions such as typhoons, the mooring damping can be adjusted by the active damping control unit to reduce the motion response of the floating platform and prevent the floating platform from capsizing.

[0103] In addition, compared to existing technologies that use fixed-length mooring cables and cannot dynamically adjust the length of the mooring cables based on real-time marine environmental data, this method uses a mooring adjustment module to calculate the wind turbine motion response, mooring tension, and wave height time history for each sea state using SIMA (Systems Integration Modeling and Analysis) software, given the known types of sea conditions, and uses this data as model training data. Using wave height time history and motion response and tension at time t as input, and response and tension after t+20s as target prediction output, the system determines whether the motion response of the floating wind turbine platform is within an acceptable range in the subsequent period, and whether the mooring cable tension exceeds a safety threshold or falls below a set value in the subsequent period. When either of these two conditions is not met, the control module sends an adjustment command to the mooring adjustment module. This process is repeated multiple times until both conditions are met. However, through multiple iterations of adjustment by the mooring adjustment module, the motion response of the floating wind turbine platform is made to be within an acceptable range in the subsequent period, and the mooring cable tension does not exceed a safety threshold or fall below a set value in the subsequent period. This increases the motion response of the floating wind turbine platform throughout its entire life cycle and cannot accurately match the mooring requirements under different sea conditions. In this embodiment, a sea state classification model is constructed by deeply mining and utilizing historical sea state information from the wind farm to classify the current sea state information. An optimal mooring cable length is pre-designed for each sea state category. When the wind turbine is identified as being in a certain sea state, the mooring cable length is adjusted to the target length, the mooring cable tension is adjusted to a preset value, and the mooring damping is adjusted to the target mooring resistance. This allows the wind farm mooring system to dynamically adjust mooring cable parameters (e.g., mooring cable length, mooring cable tension, and mooring damping) based on real-time sea state information, accurately matching mooring requirements under different sea states. This effectively reduces the tension borne by the mooring cable in extreme sea states, avoiding the risk of breakage. In stable sea states, the offset of the wind turbine floating platform is precisely controlled. By enhancing the dynamic adaptability of the mooring system to the complex and changing marine environment, the motion response of the wind turbine floating platform throughout its entire life cycle is reduced, the load is lowered, and the power generation efficiency is maintained at a high level.

[0104] In an optional implementation, S2 includes:

[0105] S21. Select machine learning and deep learning algorithms that can be used for time series classification as training algorithms;

[0106] S22. Input historical sea state information into the training algorithm to obtain a sea state classification model.

[0107] In this embodiment, machine learning and deep learning algorithms that can be used for time series classification are selected as training algorithms, such as RNN (Recurrent Neural Network) and its variants, decision trees and random forests, SVM (Support Vector Machine), etc. The processed historical sea state information is input into the training algorithm for training, verification and testing to obtain the sea state classification model.

[0108] In an optional implementation, S22 includes:

[0109] S221. Divide the historical sea state information according to a preset ratio to obtain a training set, a validation set, and a test set;

[0110] In this embodiment, for example, historical sea state information is divided into a training set, a validation set, and a test set in a ratio of 7:2:1. That is, 70% of the historical sea state information is used as the training set, 20% as the validation set, and 10% as the test set.

[0111] In a specific example, historical sea state information for the past 10 years (e.g., from 2014 to 2023) of the sea area where the wind farm is located can be downloaded from NOAA. The historical sea state information from 2014 to 2020 can be used as the training set, the historical sea state information from 2021 to 2022 can be used as the validation set, and the historical sea state information from 2023 can be used as the test set.

[0112] S222. Input the training set into the training algorithm to obtain the sea state classification model;

[0113] S223. Input the validation set into the sea state classification model to adjust the parameters of the sea state classification model based on the validation results;

[0114] In this embodiment, the sea state classification model is validated using a rolling validation method. Specifically, rolling validation uses a fixed-length time window and "rolls" forward on the time axis. Each roll performs a complete "training-validation" cycle. For example, the rolling window is 1 year of historical sea state information (e.g., historical sea state information from 2022) to perform rolling validation on the sea state classification model. The stability of the sea state classification model can be evaluated through rolling validation.

[0115] S224. Input the test set into the sea state classification model and test the sea state classification results of the sea state classification model in order to optimize the sea state classification model.

[0116] In an optional implementation, S224 includes:

[0117] Obtain the actual sea state category and the target sea state category corresponding to the test set;

[0118] The loss function is obtained based on the target sea state category and the actual sea state category;

[0119] Optimize the sea state classification model based on the loss function.

[0120] In this implementation, the imbalance in the sea state classification model is addressed using Focal Loss: a focusing parameter γ is introduced into the loss calculation, applying a (1-pt) weight to easily classified samples (high-confidence current sea state). γ The suppression weights retain higher weights for difficult-to-classify samples (current sea state with low confidence). Experiments show that when γ=2, the loss contribution of easily classified samples with a confidence of about 0.99 can be compressed by about two orders of magnitude, making the sea state classification model pay more attention to difficult-to-classify samples, thereby improving the overall classification performance.

[0121] Furthermore, by combining an optimizer with a weight decay of 0.01 (such as AdamW), the complexity of the sea state classification model is controlled by regularization, and an adaptive learning rate strategy is used to stabilize the training process and prevent overfitting.

[0122] In this embodiment, the sea state classification model is optimized by combining a loss function with an optimizer to obtain the best-performing sea state classification model, thereby improving the classification accuracy, recall, and generalization ability of the sea state classification model.

[0123] In this embodiment, S2 further includes:

[0124] At least one of the following methods is applied to historical sea state information: timestamp alignment, spatial interpolation, coordinate transformation, outlier detection, and missing value imputation, to obtain preprocessed historical sea state information.

[0125] In this embodiment, historical sea state information is preprocessed, including at least one of the following: data timestamp alignment, spatial interpolation as needed, coordinate transformation, outlier detection, and missing value imputation.

[0126] In an optional implementation, the optimization method further includes:

[0127] Store the sea state classification model and the sea state-mooring cable parameter database to the cloud network or the wind turbine side.

[0128] This implementation method utilizes historical sea state information to construct a sea state classification model. Combined with a sea state-mooring cable parameter database containing the correspondence between sea state categories and mooring cable parameters, it can dynamically adjust mooring cable parameters based on real-time sea state information, accurately match mooring requirements under different sea states, reduce fatigue load and levelized cost of electricity (LCOE) of the wind turbine floating platform, extend the service life of the wind turbine, and increase power generation.

[0129] Example 2

[0130] Corresponding to the aforementioned embodiment of the optimization method for a wind farm mooring system, this disclosure also provides an embodiment of an optimization device for a wind farm mooring system.

[0131] Figure 2 This is a schematic diagram of a module for an optimization device for a wind farm mooring system provided in Embodiment 2 of this disclosure, as shown below. Figure 2 As shown, the optimization device includes:

[0132] The first acquisition module 21 is used to acquire historical sea condition information of the wind farm;

[0133] In this embodiment, based on the location of the wind farm, historical sea state information of the sea area where the wind farm is located over the past several years (e.g., the past 10 years) is downloaded from NOAA;

[0134] In one optional implementation, the historical sea state information includes at least one of historical wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current speed information, and ocean current direction information.

[0135] The first construction module 22 is used to construct a sea state classification model based on historical sea state information;

[0136] In this embodiment, the network structure of the sea state classification model can be a CNN-LSTM network. The input layer of the CNN-LSTM network is a 7-dimensional temporal vector (time window = 60 min, stride = 15 min), the CNN convolution kernel size is 5*1, the LSTM unit is 256, the network has 3 fully connected layers (Dropout = 0.3), and the activation function is softmax = 4.

[0137] The second acquisition module 23 is used to acquire the corresponding mooring cable parameters based on the sea state category output by the sea state classification model.

[0138] The second construction module 24 is used to construct a sea state-mooring cable parameter database based on sea state categories and mooring cable parameters. The sea state-mooring cable parameter database contains at least the correspondence between sea state categories and mooring cable parameters.

[0139] In this embodiment, the mooring cable parameters include the mooring cable length and mooring damping;

[0140] In a specific example, taking mooring cable length as an example, the most severe operating parameters for each sea state category—wind, waves, and current—are selected to prepare for the design of the mooring cable length. For instance, the sea state classification model outputs four sea state categories (C1, C2, C3, and C4). Numerical simulation software (such as SIMA and AQWA) is used to design the corresponding mooring cable lengths. Following standard procedures, the mooring cable lengths corresponding to each sea state category are calculated based on water depth and the floating platform. Specifically, a set of "preset tension" and "delivery length" for a set of mooring cables are initially assumed based on sea state, platform, water depth, and mooring cable properties. Under the action of "average environmental load," the floating platform will shift, and the floating platform's displacement is calculated. When the platform is at a certain offset position, the tension provided by all mooring cables is used to further determine the equilibrium position of the floating platform, ensuring that the sum of the tensions of all mooring cables is equal in magnitude and opposite in direction to the average environmental load, and that the torques are balanced. Then, the average offset of the floating platform under this sea state category, the top tension, angle, and bottom tension of each mooring cable, as well as the seabed morphology of the mooring cables are output. The average offset of the floating platform under this sea state category, the top tension, angle, and bottom tension of each mooring cable, and the seabed morphology of the mooring cables are checked to see if they all meet the specifications and safety requirements. If at least one is not met, the "preset tension" or "cable length" is adjusted. If it is met, the mooring cable length corresponding to this sea state category is output.

[0141] As shown in Table 1, calculate the mooring cable lengths L1, L2, L3, and L4 corresponding to the four sea state categories C1, C2, C3, and C4, respectively. It should be noted that the mooring cable length is the length that minimizes the motion response of the floating platform and provides sufficient restoring force for mooring.

[0142] Table 1

[0143] category mooring cable length C1 L1 C2 L1 C3 L3 C4 L4

[0144] As shown in Table 1, each sea state category has a corresponding mooring cable length, meaning there is a correspondence between sea state categories and mooring cable lengths. Furthermore, the correspondence between sea state categories and mooring cable parameters is stored in the sea state-mooring cable parameter database.

[0145] The third acquisition module 25 is used to acquire the current sea state information of the wind farm;

[0146] In an optional implementation, the third acquisition module is used to acquire at least one of the following information based on the data acquisition equipment on the wind farm or wind turbine side: current wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current velocity information, and ocean current direction information. The data acquisition equipment includes at least one of the following: a wind measurement tower, an acoustic Doppler current meter, and a laser wave meter.

[0147] In this embodiment, wind speed and wind direction information of the wind farm are measured by a wind measuring tower, ocean current velocity and ocean current direction information are measured by an acoustic Doppler current meter, and wave height, wave period and wave direction information are measured by a laser wave meter.

[0148] The fourth acquisition module 26 is used to input the current sea state information into the sea state classification model to obtain the current sea state category corresponding to the current sea state information;

[0149] The fifth acquisition module 27 is used to acquire the target mooring cable parameters corresponding to the current sea state category based on the sea state-mooring cable parameter database.

[0150] In this embodiment, the current sea state information is input into the sea state classification model to obtain the current sea state category corresponding to the current sea state information; further, the target mooring cable parameters corresponding to the current sea state information are obtained based on the sea state-mooring cable parameter database.

[0151] In an optional implementation, the optimization device further includes:

[0152] The sixth acquisition module is used to acquire the current mooring cable parameters corresponding to the current sea state category;

[0153] The adjustment module is used to adjust the current mooring cable parameters based on the target mooring cable parameters in response to the inconsistency between the current mooring cable parameters and the target mooring cable parameters, so as to make the current mooring cable parameters consistent with the target mooring cable parameters.

[0154] In an optional implementation, the mooring cable parameters include mooring cable length and mooring damping. An adjustment module is used to adjust the current mooring cable length and current mooring damping through a mooring adjustment unit so that the current mooring cable length is consistent with the target mooring cable length and the current mooring damping is consistent with the target mooring damping.

[0155] In this embodiment, when the current mooring cable length is inconsistent with the target mooring cable length and / or the current mooring damping is inconsistent with the target mooring damping, the wind farm mooring system activates the mooring adjustment unit to adjust the current mooring cable length and the current mooring damping so that the current mooring cable length and the current mooring damping are consistent with the target mooring cable length and the target mooring damping.

[0156] It should be noted that in this embodiment, the length of the mooring cable and the mooring damping can be adjusted by the mooring adjustment unit (for example, the mooring adjustment unit can be a conventional tensioner, anchor winch, chain stopper, etc.). Alternatively, the wind farm mooring system can be equipped with an active damping control unit to adjust the mooring damping according to different sea conditions and the length of the mooring cable. For example, in the case of severe sea conditions such as typhoons, the mooring damping can be adjusted by the active damping control unit to reduce the motion response of the floating platform and prevent the floating platform from capsizing.

[0157] Furthermore, compared to existing technologies that use fixed-length mooring cables, which cannot dynamically adjust the length based on real-time marine environmental data; or those that use a mooring adjustment module, given the known sea conditions, to calculate the wind turbine's motion response, mooring tension, and wave height time history using SIMA software and use this as model training data, this approach uses wave height time history and the motion response and tension at time t as input, and the response and tension after t+20s as the target prediction output. It determines whether the wind turbine floating platform's motion response remains within an acceptable range in the subsequent period, and whether the mooring cable tension exceeds a safety threshold or falls below a set value in the subsequent period. If either of these conditions is not met, the control module sends an adjustment command to the mooring adjustment module, iterating multiple times until both conditions are met. However, while the mooring adjustment module achieves acceptable motion response and ensures the mooring cable tension does not exceed a safety threshold or fall below a set value in the subsequent period through multiple iterations, this increases the overall motion response of the wind turbine floating platform throughout its lifecycle and cannot accurately match the mooring requirements under different sea conditions. In this embodiment, a sea state classification model is constructed by deeply mining and utilizing historical sea state information from the wind farm to classify the current sea state information. An optimal mooring cable length is pre-designed for each sea state category. When the wind turbine is identified as being in a certain sea state, the mooring cable length is adjusted to the target length, the mooring cable tension is adjusted to a preset value, and the mooring damping is adjusted to the target mooring resistance. This allows the wind farm mooring system to dynamically adjust mooring cable parameters (e.g., mooring cable length, mooring cable tension, and mooring damping) based on real-time sea state information, accurately matching mooring requirements under different sea states. This effectively reduces the tension borne by the mooring cable in extreme sea states, avoiding the risk of breakage. In stable sea states, the offset of the wind turbine floating platform is precisely controlled. By enhancing the dynamic adaptability of the mooring system to the complex and changing marine environment, the motion response of the wind turbine floating platform throughout its entire life cycle is reduced, the load is lowered, and the power generation efficiency is maintained at a high level.

[0158] In an optional implementation, the first building module includes:

[0159] The selection unit is used to select machine learning and deep learning algorithms that can be used for time series classification as training algorithms;

[0160] The acquisition unit is used to input historical sea state information into the training algorithm to obtain a sea state classification model.

[0161] In this embodiment, machine learning and deep learning algorithms that can be used for time series classification are selected as training algorithms, such as RNN (Recurrent Neural Network) and its variants, decision trees and random forests, SVM (Support Vector Machine), etc. The processed historical sea state information is input into the training algorithm for training, verification and testing to obtain the sea state classification model.

[0162] In an optional implementation, the acquisition unit includes:

[0163] The data is divided into sub-units to divide historical sea state information according to a preset ratio to obtain training set, validation set and test set;

[0164] In this embodiment, for example, historical sea state information is divided into a training set, a validation set, and a test set in a ratio of 7:2:1. That is, 70% of the historical sea state information is used as the training set, 20% as the validation set, and 10% as the test set.

[0165] In a specific example, historical sea state information for the past 10 years (e.g., from 2014 to 2023) of the sea area where the wind farm is located can be downloaded from NOAA. The historical sea state information from 2014 to 2020 can be used as the training set, the historical sea state information from 2021 to 2022 can be used as the validation set, and the historical sea state information from 2023 can be used as the test set.

[0166] Obtain sub-units to input the training set into the training algorithm to obtain the sea state classification model;

[0167] The adjustment sub-unit is used to input the validation set into the sea state classification model in order to adjust the parameters of the sea state classification model based on the validation results;

[0168] In this embodiment, the sea state classification model is validated using a rolling validation method. Specifically, rolling validation uses a fixed-length time window and "rolls" forward on the time axis. Each roll performs a complete "training-validation" cycle. For example, the rolling window is 1 year of historical sea state information (e.g., historical sea state information from 2022) to perform rolling validation on the sea state classification model. The stability of the sea state classification model can be evaluated through rolling validation.

[0169] The test unit is used to input the test set into the sea state classification model, test the sea state classification results of the sea state classification model, and optimize the sea state classification model.

[0170] In one optional implementation, the testing unit is used to obtain the real sea state category and the target sea state category corresponding to the test set; obtain a loss function based on the target sea state category and the real sea state category; and optimize the sea state classification model based on the loss function.

[0171] In this implementation, Focal Loss is used to address the imbalance in the sea state classification model:

[0172] A focusing parameter γ is introduced into the loss calculation, applying a (1-pt) pressure to easily classifiable samples (high-confidence current sea state). γ The suppression weights retain higher weights for difficult-to-classify samples (current sea state with low confidence). Experiments show that when γ=2, the loss contribution of easily classified samples with a confidence of about 0.99 can be compressed by about two orders of magnitude, making the sea state classification model pay more attention to difficult-to-classify samples, thereby improving the overall classification performance.

[0173] Furthermore, by combining an optimizer with a weight decay of 0.01 (such as AdamW), the complexity of the sea state classification model is controlled by regularization, and an adaptive learning rate strategy is used to stabilize the training process and prevent overfitting.

[0174] In this implementation, the sea state classification model is optimized by combining a loss function with an optimizer to obtain the best-performing sea state classification model, thereby improving the classification accuracy, recall, and generalization ability of the sea state classification model.

[0175] In this embodiment, the first building module further includes:

[0176] The preprocessing unit is used to perform at least one of the following on historical sea state information: timestamp alignment, spatial interpolation, coordinate transformation, outlier detection, and missing value imputation, to obtain preprocessed historical sea state information.

[0177] In this embodiment, historical sea state information is preprocessed, including at least one of the following: data timestamp alignment, spatial interpolation as needed, coordinate transformation, outlier detection, and missing value imputation.

[0178] In an optional implementation, the optimization device further includes:

[0179] The storage module is used to store the sea state classification model and the sea state-mooring cable parameter database to the cloud network or the wind turbine side.

[0180] This implementation method utilizes historical sea state information to construct a sea state classification model. Combined with a sea state-mooring cable parameter database containing the correspondence between sea state categories and mooring cable parameters, it can dynamically adjust mooring cable parameters based on real-time sea state information, accurately match mooring requirements under different sea states, reduce fatigue load and levelized cost of electricity (LCOE) of the wind turbine floating platform, extend the service life of the wind turbine, and increase power generation.

[0181] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs.

[0182] Example 3

[0183] Figure 3 This is a schematic diagram of the structure of an electronic device according to Embodiment 3 of this disclosure. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the optimization method of the wind farm mooring system described in any of the above embodiments. Figure 3 The electronic device 90 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0184] like Figure 3 As shown, the electronic device 90 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 90 may include, but are not limited to: at least one processor 91, at least one memory 92, and a bus 93 connecting different system components (including memory 92 and processor 91).

[0185] Bus 93 includes a data bus, an address bus, and a control bus.

[0186] The memory 92 may include volatile memory, such as random access memory (RAM) 921 and / or cache memory 922, and may further include read-only memory (ROM) 923.

[0187] The memory 92 may also include a program tool 925 (or utility) having a set (at least one) program module 924, such program module 924 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0188] The processor 91 executes various functional applications and data processing by running computer programs stored in the memory 92, such as the optimization method for wind farm mooring systems provided in any of the above embodiments.

[0189] Electronic device 90 can also communicate with one or more external devices 94 (e.g., keyboard, pointing device, etc.). This communication can be performed via input / output (I / O) interface 95. Furthermore, electronic device 90 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 96. Figure 3 As shown, network adapter 96 communicates with other modules of electronic device 90 via bus 93. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0190] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0191] Example 4

[0192] Embodiment 4 of this disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the optimization method for the wind farm mooring system provided in any of the above embodiments.

[0193] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0194] Example 5

[0195] Embodiment 5 of this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the optimization method for the wind farm mooring system described in any of the preceding claims.

[0196] The program code for executing the computer program product of this disclosure can be written in any combination of one or more programming languages, and the program code can be executed entirely on a user device, partially on a user device, as a stand-alone software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0197] While specific embodiments of this disclosure have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of this disclosure is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of this disclosure, but all such changes and modifications fall within the scope of protection of this disclosure.

Claims

1. An optimization method for a wind farm mooring system, characterized in that, The optimization method includes: Obtain historical sea condition information for wind farms; A sea state classification model is constructed based on the historical sea state information; Based on the sea state category output by the sea state classification model, obtain the corresponding mooring cable parameters; A sea state-mooring cable parameter database is constructed based on the sea state categories and mooring cable parameters. The sea state-mooring cable parameter database contains at least the correspondence between sea state categories and mooring cable parameters. Obtain current sea state information for the wind farm; The current sea state information is input into the sea state classification model to obtain the current sea state category corresponding to the current sea state information; Based on the sea state-mooring cable parameter database, obtain the target mooring cable parameters corresponding to the current sea state category.

2. The optimization method for a wind farm mooring system as described in claim 1, characterized in that, The optimization method further includes: Obtain the current mooring cable parameters corresponding to the current sea state category; In response to the inconsistency between the current mooring cable parameters and the target mooring cable parameters, the current mooring cable parameters are adjusted based on the target mooring cable parameters to make the current mooring cable parameters consistent with the target mooring cable parameters.

3. The optimization method for a wind farm mooring system as described in claim 2, characterized in that, The mooring cable parameters include mooring cable length and mooring damping. Adjusting the current mooring cable parameters based on the target mooring cable parameters to ensure consistency between the current mooring cable parameters and the target mooring cable parameters includes: The current mooring cable length and the current mooring damping are adjusted by the mooring adjustment unit to make the current mooring cable length consistent with the target mooring cable length and the current mooring damping consistent with the target mooring damping.

4. The optimization method for a wind farm mooring system as described in claim 1, characterized in that, The sea state classification model constructed based on the historical sea state information includes: Machine learning and deep learning algorithms suitable for time series classification were selected as training algorithms; The historical sea state information is input into the training algorithm to obtain the sea state classification model.

5. The optimization method for a wind farm mooring system as described in claim 4, characterized in that, The step of inputting the historical sea state information into the training algorithm to obtain the sea state classification model includes the following steps: The historical sea state information is divided according to a preset ratio to obtain a training set, a validation set, and a test set; The training set is input into the training algorithm to obtain the sea state classification model; The validation set is input into the sea state classification model to adjust the parameters of the sea state classification model based on the validation results; The test set is input into the sea state classification model to test the sea state classification results of the sea state classification model, so as to optimize the sea state classification model.

6. The optimization method for a wind farm mooring system as described in claim 5, characterized in that, The step of inputting the test set into the sea state classification model and testing the sea state classification results of the sea state classification model in order to optimize the sea state classification model includes: Obtain the actual sea state category and the target sea state category corresponding to the test set; A loss function is obtained based on the target sea state category and the actual sea state category; The sea state classification model is optimized based on the loss function.

7. The optimization method for a wind farm mooring system as described in claim 1, characterized in that, The sea state classification model constructed based on the historical sea state information also includes: The historical sea state information is processed by at least one of the following methods: timestamp alignment, spatial interpolation, coordinate transformation, outlier detection, and missing value imputation, to obtain preprocessed historical sea state information.

8. The optimization method for a wind farm mooring system as described in claim 1, characterized in that, The historical sea condition information includes at least one of the following: historical wind speed information, wind direction information, wave height information, wave period information, wave direction information, ocean current speed information, and ocean current direction information.

9. The optimization method for a wind farm mooring system as described in claim 1, characterized in that, The optimization method further includes: The sea state classification model and the sea state-mooring cable parameter database are stored in the cloud network or on the wind turbine side.

10. The optimization method for a wind farm mooring system as described in claim 1, characterized in that, The acquisition of the current sea state information of the wind farm includes: Based on the data acquisition equipment on the wind farm or wind turbine side, at least one of the following information is collected: current wind speed, wind direction, wave height, wave period, wave direction, ocean current velocity, and ocean current direction. The data acquisition equipment includes at least one of the following: a wind measurement tower, an acoustic Doppler current meter, and a laser wave meter.

11. An optimization device for a wind farm mooring system, characterized in that, The optimization device includes: The first acquisition module is used to acquire historical sea condition information of the wind farm; The first construction module is used to construct a sea state classification model based on the historical sea state information; The second acquisition module is used to acquire the corresponding mooring cable parameters based on the sea state category output by the sea state classification model. The second construction module is used to construct a sea state-mooring cable parameter database based on the sea state category and mooring cable parameters. The sea state-mooring cable parameter database contains at least the correspondence between sea state categories and mooring cable parameters. The third acquisition module is used to acquire the current sea state information of the wind farm; The fourth acquisition module is used to input the current sea state information into the sea state classification model to obtain the current sea state category corresponding to the current sea state information; The fifth acquisition module is used to acquire the target mooring cable parameters corresponding to the current sea state category based on the sea state-mooring cable parameter database.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the optimization method for the wind farm mooring system as described in any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the optimization method for the wind farm mooring system as described in any one of claims 1 to 10.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the optimization method for the wind farm mooring system as described in any one of claims 1 to 10.

Images