Method and device for determining fatigue loss of pile foundation column of offshore photovoltaic, and medium

Abstract

This disclosure relates to a method, apparatus, and medium for determining fatigue loss of pile foundation columns for offshore photovoltaic systems, belonging to the field of fatigue analysis technology. The method includes: determining dynamic data of pile foundation loads on the pile foundation columns based on historical column status data; determining the target remaining life of a three-dimensional digital twin model under extreme service conditions based on the dynamic data of pile foundation loads; training a neural network model to obtain a fatigue life prediction model based on the target remaining life, column material parameters, dynamic data of pile foundation loads, and historical environmental variable data corresponding to the historical column status data; inputting the real-time model status data of the three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and determining the real-time fatigue loss of the pile foundation columns based on the real-time remaining life probability distribution. This disclosure improves the accuracy of fatigue loss analysis results.

Description

Technical Field

[0001] This disclosure relates to the field of fatigue analysis technology, and more specifically, to a method for determining the fatigue loss of the pile foundation column of an offshore photovoltaic system, a device for determining the fatigue loss of the pile foundation column of an offshore photovoltaic system, and a computer-readable storage medium. Background Technology

[0002] The existing solutions do not mention how to analyze the real-time fatigue damage of the pile foundation columns.

[0003] It should be noted that the information in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this disclosure is to provide a method for determining the fatigue loss of the pile foundation column of an offshore photovoltaic system, a device for determining the fatigue loss of the pile foundation column of an offshore photovoltaic system, and a computer-readable storage medium, thereby overcoming, to at least a certain extent, the problem that it is impossible to analyze the real-time fatigue damage of the pile foundation column due to the limitations and defects of related technologies.

[0005] According to one aspect of this disclosure, a method for determining fatigue loss of a pile foundation column for offshore photovoltaic (PV) systems is provided, comprising: acquiring historical column status data of the pile foundation column in an offshore PV scenario, and determining dynamic data of the pile foundation load of the pile foundation column based on the historical column status data; wherein the dynamic data of the pile foundation load is obtained by: acquiring buoys, anemometers, and displacement sensors corresponding to the pile foundation column in an offshore PV scenario, and acquiring historical wave buoy data monitored by the buoys, historical wind speed and direction data monitored by the anemometers, and historical pile foundation strain data monitored by the displacement sensors; determining historical ocean energy spectrum data of the pile foundation column based on the historical wave buoy data, and determining historical three-dimensional turbulent wind field data of the pile foundation column based on the historical wind speed and direction data; and determining the pile foundation strain based on the historical pile foundation strain data. Historical fluid-structure interaction data of the pile foundation column is used, and dynamic data of the pile foundation load of the pile foundation column is constructed based on the historical ocean energy spectrum data, historical three-dimensional turbulent wind field data, and historical fluid-structure interaction data. A three-dimensional digital twin model corresponding to the pile foundation column is constructed, and the target remaining life of the three-dimensional digital twin model under extreme service conditions is determined based on the dynamic data of the pile foundation load. Based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column state data, a neural network model to be trained is trained to obtain a fatigue life prediction model. The real-time model state data of the three-dimensional digital twin model is input into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and the real-time fatigue loss of the pile foundation column is determined based on the real-time remaining life probability distribution.

[0006] In one exemplary embodiment of this disclosure, determining the historical ocean energy spectrum data of the pile foundation column based on the historical wave buoy data includes: determining a wave spectrum function, determining a spectrum diagram based on the wave spectrum function, and determining the average period of the wave's upward span based on the spectrum diagram. Average period of spectral peaks Among them, the upward average period Average period of spectral peaks The following relationship exists between them: ;in, For spectral coefficients; based on the average period of the upper span Average period of spectral peaks and the upward average period Average period of spectral peaks The relationship between the two factors determines the average period. The specific value of the average period is: Based on the upward average period Average period of spectral peaks Average period The height of the waves was used to determine the historical ocean energy spectrum data of the pile foundation columns.

[0007] In one exemplary embodiment of this disclosure, constructing a three-dimensional digital twin model corresponding to the pile foundation column includes: obtaining an original geometric model of the pile foundation column under an offshore photovoltaic scenario, and configuring first component performance parameters for the original geometric model; wherein, the first component performance parameters include at least one of a first elastic modulus, a first Poisson's ratio, a first tensile strength, a first compressive strength, and a first horizontal bearing capacity of the first component material used to prepare the pile foundation column; applying a first simulated service load under simulated service conditions to the original geometric model; wherein, the simulated service conditions are obtained by simulating the actual service conditions of the pile foundation column, and the first simulated service load includes at least one of a first weather scenario load, a first pressure load, and a first support force load borne by the pile foundation column during service; and generating a three-dimensional digital twin model corresponding to the pile foundation column based on the original geometric model, the first component performance parameters, and the first simulated service load.

[0008] In an exemplary embodiment of this disclosure, determining the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load includes: determining the fatigue damage coefficient of the pile foundation column due to the combined effects of waves, tides, and sea winds based on the dynamic data of the pile foundation load and historical environmental variable data corresponding to historical column status data; determining the critical stress area of ​​the three-dimensional digital twin model and determining the original remaining life of the critical stress area under extreme service conditions based on the fatigue damage coefficient; determining the cumulative damage of the critical stress area and the crack propagation path caused by the cumulative damage, and correcting the original remaining life based on the crack propagation path to obtain the target remaining life of the three-dimensional digital twin model under extreme service conditions.

[0009] In one exemplary embodiment of this disclosure, a fatigue life prediction model is obtained by training a neural network model to be trained based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data. This includes: constructing a first original feature map based on the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data; inputting the first original feature map into the neural network model to be trained to obtain a fatigue life prediction result for the pile foundation column; constructing a loss function based on the fatigue life prediction result and the target remaining life, and training the neural network model to be trained based on the loss function to obtain the fatigue life prediction model.

[0010] In one exemplary embodiment of this disclosure, a first original feature map is constructed based on the column material parameters of the pile foundation column, dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column state data. This includes: using the pile foundation column in an offshore photovoltaic scenario as a first vertex, and the connection relationship between the pile foundation columns as a first connecting edge; determining the first edge weight of the first connecting edge based on the positional relationship between the pile foundation columns, and using the column material parameters of the pile foundation column, dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column state data as a second vertex, a third vertex, and a fourth vertex; determining the second connecting edge, a third connecting edge, and a fourth connecting edge based on the relationship between the pile foundation column and the column material parameters, dynamic data of the pile foundation load, and historical environmental variable data; and constructing the first original feature map based on the first vertex, the second vertex, the third vertex, the fourth vertex, the first connecting edge, the second connecting edge, the third connecting edge, the fourth connecting edge, and the first edge weight.

[0011] In an exemplary embodiment of this disclosure, the fatigue life prediction model includes a first graph convolutional layer, a second graph convolutional layer, ..., an Nth graph convolutional layer, and a classification layer. The process of inputting real-time model state data of a three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model includes: constructing a second original feature map based on the real-time model state data of the three-dimensional digital twin model and the column material parameters of the three-dimensional digital twin model; updating the node features of the second original feature map based on the first graph convolutional layer to obtain a first graph convolutional processing result; updating the node features of the first graph convolutional processing result based on the second graph convolutional layer to obtain a second graph convolutional processing result; and repeating the determination process of the second graph convolutional processing result sequentially to obtain a third graph convolutional processing result, ..., an Nth graph convolutional processing result; and classifying the Nth graph convolutional processing result based on the classification layer to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model.

[0012] According to one aspect of this disclosure, a device for determining the fatigue loss of a pile foundation column for offshore photovoltaic systems is provided, comprising: a pile foundation load dynamic data determination module, used to acquire historical column state data of the pile foundation column in an offshore photovoltaic scenario, and determine the pile foundation load dynamic data of the pile foundation column based on the historical column state data; a target remaining life determination module, used to construct a three-dimensional digital twin model corresponding to the pile foundation column, and determine the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the pile foundation load dynamic data; a fatigue life prediction model determination module, used to train a neural network model to be trained based on the target remaining life, the column material parameters of the pile foundation column, the pile foundation load dynamic data, and historical environmental variable data corresponding to the historical column state data to obtain a fatigue life prediction model; and a real-time fatigue loss determination module, used to input the real-time model state data of the three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and determine the real-time fatigue loss of the pile foundation column based on the real-time remaining life probability distribution.

[0013] According to one aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for determining fatigue loss of the pile foundation column of offshore photovoltaic systems as described in any of the preceding claims.

[0014] According to one aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the fatigue loss determination method for the foundation columns of offshore photovoltaic systems according to any one of the preceding claims by executing the executable instructions.

[0015] This disclosure provides a method for determining the fatigue loss of pile foundation columns in offshore photovoltaic systems. Firstly, it acquires historical column status data of the pile foundation columns in an offshore photovoltaic scenario and determines the dynamic data of the pile foundation load based on this data. Then, it constructs a three-dimensional digital twin model corresponding to the pile foundation columns and determines the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load. Next, based on the target remaining life, the column material parameters, the dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column status data, it trains a neural network model to obtain a fatigue life prediction model. Finally, it inputs the real-time model status data of the three-dimensional digital twin model into the fatigue life prediction model. This method obtains the real-time remaining life probability distribution of a 3D digital twin model and determines the real-time fatigue loss of the pile foundation column based on this distribution. This enables real-time analysis of the fatigue loss of the pile foundation column, solving the problem that existing technologies cannot perform real-time analysis of the fatigue loss of pile foundation columns. Furthermore, because the real-time model state data of the 3D digital twin model can be input into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the 3D digital twin model, the accuracy of the obtained real-time remaining life probability distribution is improved. Moreover, because the fatigue loss of the pile foundation column can be analyzed in real-time, the dangerous columns can be repaired promptly based on the analysis results, thus avoiding safety problems caused by pile foundation column damage.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0018] Figure 1 The flowchart schematically illustrates a method for determining fatigue loss of a pile foundation column for offshore photovoltaic systems according to an exemplary embodiment of the present disclosure.

[0019] Figure 2 The diagram schematically illustrates a structural example of a fatigue life prediction model obtained according to an exemplary embodiment of the present disclosure.

[0020] Figure 3 The diagram schematically illustrates a structural example of a large model generated from content used in generating a three-dimensional digital twin model according to an exemplary embodiment of this disclosure.

[0021] Figure 4 The diagram schematically illustrates an example structure of a hybrid expert model layer in a content generation large model according to an example embodiment of the present disclosure.

[0022] Figure 5 The illustration schematically shows a scenario example diagram of a first original feature map obtained according to an exemplary embodiment of the present disclosure.

[0023] Figure 6 The diagram illustrates a scenario example of a fatigue life prediction process according to an exemplary embodiment of the present disclosure.

[0024] Figure 7 The diagram schematically illustrates a structural example of a device for determining fatigue loss of a pile foundation column for offshore photovoltaic systems according to an exemplary embodiment of the present disclosure.

[0025] Figure 8 An electronic device is illustrated in accordance with an example embodiment of the present disclosure for a method of determining fatigue loss of a pile foundation column for realizing offshore photovoltaics. Detailed Implementation

[0026] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this disclosure more comprehensive and complete, and to fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0027] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0028] This exemplary embodiment first provides a method for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems. This method can run on terminal devices, servers, server clusters, or cloud servers, etc. Of course, those skilled in the art can also run the method disclosed herein on other platforms as needed, and this exemplary embodiment does not impose any special limitations on this. Specifically, refer to... Figure 1 As shown, the method for determining the fatigue loss of the pile foundation column of the offshore photovoltaic system may include the following steps:

[0029] Step S110. Obtain historical column status data of the pile foundation columns in the offshore photovoltaic scenario, and determine the dynamic data of the pile foundation load of the pile foundation columns based on the historical column status data;

[0030] Step S120. Construct a three-dimensional digital twin model corresponding to the pile foundation column, and determine the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load;

[0031] Step S130. Based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data, train the neural network model to be trained to obtain a fatigue life prediction model.

[0032] Step S140. Input the real-time model state data of the three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and determine the real-time fatigue loss of the pile column based on the real-time remaining life probability distribution.

[0033] In the aforementioned method for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems, the following steps are taken: First, historical column status data of the pile foundation columns in the offshore photovoltaic scenario is acquired, and dynamic data of the pile foundation load is determined based on this data. Then, a three-dimensional digital twin model corresponding to the pile foundation column is constructed, and the target remaining life of the three-dimensional digital twin model under extreme service conditions is determined based on the dynamic data of the pile foundation load. Next, based on the target remaining life, the column material parameters, the dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column status data, a neural network model is trained to obtain a fatigue life prediction model. Finally, the real-time model status data of the three-dimensional digital twin model is input into the fatigue life prediction model. This method obtains the real-time remaining life probability distribution of a 3D digital twin model and determines the real-time fatigue loss of pile foundation columns based on this distribution. This enables real-time analysis of fatigue loss in pile foundation columns, solving the problem that existing technologies cannot perform real-time analysis of fatigue loss in pile foundation columns. Furthermore, because the real-time model state data of the 3D digital twin model can be input into the fatigue life prediction model to obtain the real-time remaining life probability distribution, the accuracy of the obtained real-time remaining life probability distribution is improved. Moreover, because the fatigue loss of pile foundation columns can be analyzed in real-time, the analysis results allow for timely repair of dangerous columns, thus preventing safety issues caused by pile foundation column damage.

[0034] The following will further explain and illustrate the method for determining fatigue loss of the pile foundation column of offshore photovoltaic system as described in the exemplary embodiments of this disclosure, with reference to the accompanying drawings.

[0035] First, the application scenarios of the exemplary embodiments of this disclosure will be explained and described. Specifically, the fatigue loss determination method for the pile foundation columns of offshore photovoltaic systems described in the exemplary embodiments of this disclosure can be used to analyze the real-time fatigue loss of the pile foundation columns in offshore photovoltaic scenarios, and when it is determined that the real-time fatigue loss value of any pile foundation column reaches a critical threshold, corresponding measures (such as replacement or repair) can be taken in a timely manner to avoid safety problems and economic losses caused by the damage of the pile foundation columns.

[0036] Secondly, the neural network model to be trained involved in the exemplary embodiments of this disclosure will be explained and described. Specifically, refer to... Figure 2 As shown, the fatigue life prediction model may include a first input layer 201, multiple graph convolutional layers (e.g., first graph convolutional layer, second graph convolutional layer, ..., Nth graph convolutional layer) 202, a classification layer 203, and a first output layer 204; the roles of each model layer in the fatigue life prediction process will be detailed later, and will not be elaborated further here.

[0037] The following will be about Figure 1 The method for determining the fatigue loss of the pile foundation columns of the offshore photovoltaic system shown in the figure will be further explained and illustrated.

[0038] In step S110, historical column status data of the pile foundation column under the offshore photovoltaic scenario is obtained, and the dynamic data of the pile foundation load of the pile foundation column is determined based on the historical column status data.

[0039] Specifically, the historical column status data recorded here may include, but is not limited to, historical wave buoy data, historical wind speed and direction data, and historical pile foundation strain data. In actual applications, these can be selected according to actual needs, and this example does not impose any special restrictions. Furthermore, in actual offshore photovoltaic scenarios, each pile foundation column will be equipped with a corresponding buoy, anemometer, and displacement sensor. Based on the equipped buoy, anemometer, and displacement sensor, historical wave buoy data, historical wind speed and direction data, and historical pile foundation strain data at different times will be collected. If it is necessary to obtain the historical column status data of the pile foundation columns in an offshore photovoltaic scenario, the mapping relationship between the pile foundation columns and the buoy, anemometer, and displacement sensor can be used to determine the buoy, anemometer, and displacement sensor corresponding to the pile foundation column. The historical wave buoy data monitored by the buoy, the historical wind speed and direction data monitored by the anemometer, and the historical pile foundation strain data monitored by the displacement sensor can then be obtained to obtain the historical column status data recorded above.

[0040] Secondly, the dynamic data of the pile foundation load of the pile foundation column is determined based on the historical column status data. Specifically, this can be achieved as follows: the historical ocean energy spectrum data of the pile foundation column is determined based on the historical wave buoy data, and the historical three-dimensional turbulent wind field data of the pile foundation column is determined based on the historical wind speed and direction data; the historical fluid-structure interaction data of the pile foundation column is determined based on the historical pile foundation strain data, and the dynamic data of the pile foundation load of the pile foundation column is constructed based on the historical ocean energy spectrum data, the historical three-dimensional turbulent wind field data, and the historical fluid-structure interaction data.

[0041] In one example embodiment, the historical ocean energy spectrum data of the pile foundation column is determined based on historical wave buoy data, which can be achieved as follows: First, the calculation formula for the wave spectrum is determined; specifically, the calculation formula for the wave spectrum described herein can be shown in the following formula (1):

[0042] ;Formula (1)

[0043] in, For the spectrum function, This is the actual frequency of wave fluctuations (the specific value can be determined from historical wave buoy data). The height of the wave (the specific value can be determined from historical wave buoy data). This is the reference undulation frequency of the wave. , as well as For spectral coefficients; further, The specific values ​​can be selected according to the following principles:

[0044] ;Formula (2)

[0045] Among them, when hour, ;when hour, .

[0046] at the same time, The specific values ​​should follow the following principles:

[0047] ;Formula (3)

[0048] That is, Need to ensure as well as This satisfies formula (3). Furthermore, The value can be between 1 and 10, with a standard value of 3.3.

[0049] Secondly, based on the wave spectrum function described above, the corresponding spectrum diagram can be obtained; and, based on the spectrum diagram, the average period of the wave's upward span can be calculated. Average period of spectral peaks At the same time, crossing the average cycle. Average period of spectral peaks The following relationship exists between them:

[0050] ;Formula (4)

[0051] Obtaining the upper average period Average period of spectral peaks and the upward average period Average period of spectral peaks After determining the relationship between them, the average period can be obtained. The specific value of the average period is: .

[0052] Finally, based on the upward average period Average period of spectral peaks Average period Historical ocean energy spectrum data can be obtained by measuring the height of the waves.

[0053] In one example embodiment, the historical three-dimensional turbulent wind field data of the pile foundation column can be determined based on historical wind speed and direction data in the following way: First, construct the theoretical power density spectrum formula of the longitudinal wind speed of the pile foundation column model; wherein, the power density spectrum formula can be as shown in the following formula (5):

[0054] ;Formula (5)

[0055] in, This refers to the frequency of wind (the specific value can be determined from historical wind speed and direction data). For power density, This represents the standard deviation of the actual wind speed (the specific value can be determined from historical wind speed and direction data). This is a turbulence-scale parameter, which can be determined based on the height of the pile foundation column above sea level; The wind speed at the pile foundation column is used (the specific value can be determined from historical wind speed and direction data); based on the obtained power density, wind frequency, standard deviation of actual wind speed, turbulence scale parameters, and wind speed at the pile foundation column, historical three-dimensional turbulent wind field data can be obtained.

[0056] In one example embodiment, the historical fluid-structure interaction data of the pile foundation column can be determined based on historical pile foundation strain data as follows: First, a solid field corresponding to the pile foundation column and a fluid field corresponding to the seawater are constructed; second, the pressure of the flow field is obtained by solving the fluid field based on the historical pile foundation strain data; then, the solid field is solved in the structural module (steady-state or transient) based on the historical pile foundation strain data and the solid field to obtain the deformation of the pile foundation column; further, the historical fluid-structure interaction data can be obtained based on the obtained pressure of the flow field and the deformation of the pile foundation column.

[0057] In step S120, a three-dimensional digital twin model corresponding to the pile foundation column is constructed, and the target remaining life of the three-dimensional digital twin model under extreme service conditions is determined based on the dynamic data of the pile foundation load.

[0058] In this example embodiment, firstly, a three-dimensional digital twin model corresponding to the pile foundation column is constructed. Specifically, this can be achieved as follows: The original geometric model of the pile foundation column under the offshore photovoltaic scenario is obtained, and first component performance parameters are configured for the original geometric model. The first component performance parameters include at least one of the first elastic modulus, first Poisson's ratio, first tensile strength, first compressive strength, and first horizontal bearing capacity of the first component material used to prepare the pile foundation column. A first simulated service load under simulated service conditions is applied to the original geometric model. The simulated service conditions are obtained by simulating the actual service conditions of the pile foundation column, and the first simulated service load includes at least one of the first weather scenario load, first pressure load, and first support force load borne by the pile foundation column during service. Based on the original geometric model, the first component performance parameters, and the first simulated service load, a three-dimensional digital twin model corresponding to the pile foundation column is generated. It should be noted that in generating the 3D digital twin model corresponding to the pile foundation column, it can be achieved by generating a large model based on the corresponding content, or it can be achieved based on 3D simulation software (such as ANSYS Workbench). This example does not impose any special restrictions on this.

[0059] In one example embodiment, if a 3D digital twin model is determined by generating a large content model, it can be achieved as follows: First, refer to... Figure 3 As shown, the content generation large model described here may include a second input layer 301, an embedding mapping layer 302, an encoding layer 303, a hybrid expert model layer 304, and a second output layer 305; furthermore, when generating a three-dimensional digital twin model, the original geometric model, the performance parameters of the first component, and the first simulated service load can be input into the content generation large model to obtain a three-dimensional digital twin model.

[0060] In one example embodiment, a three-dimensional digital twin model is obtained by inputting the original geometric model, the performance parameters of the first component, and the first simulated service load into a content generation large model. This can be achieved as follows: First basic information to be predicted is generated based on the original geometric model, the performance parameters of the first component, and the first simulated service load; first context information to be predicted is generated based on preset first model prompt parameters; first basic information to be predicted is processed by embedding mapping layer to obtain first model features; first context information to be predicted is processed by embedding mapping layer to obtain first context flag sequence; first model features and first context flag sequence are encoded by encoding layer to obtain first overall context representation; and first context flag sequence and first overall context representation are generated by hybrid expert model layer to obtain the three-dimensional digital twin model. The preset first model prompt parameters described here could be, for example: Your task is xxx, you need to generate a xxxx based on the input data; the generated model needs to be displayed in xxx format; the generated model can include x, ... In practical applications, corresponding first model prompt parameters can be set according to actual needs; this example does not impose special restrictions on this. Furthermore, the embedding mapping layer described herein may include an embedding embedding mapping layer and a BERT embedding mapping layer. In practical applications, the embedding embedding mapping layer can be used to perform embedding mapping processing on the first basic information to be predicted to obtain the first model features, and the BERT embedding mapping layer can be used to perform embedding mapping processing on the first context information to be predicted to obtain the first context label sequence. The encoding layer described herein may be a bidirectional multi-layer Transformer. The hybrid expert model layer described herein may include a first gated network model and multiple first expert neural network models. Specific example diagrams can be found in the provided text. Figure 4 As shown.

[0061] In one example embodiment, a three-dimensional digital twin model is obtained by generating a model based on a hybrid expert model layer using a first context flag sequence and a first overall representation of the context. This can be achieved as follows: Based on a first gated network model and the first context flag sequence, a first model weight for performing a model generation task in the model fit dimension and a second model weight for performing a model generation task in the model fitness dimension are determined by the first expert neural network model. Based on the first model weight, a first target neural network model required for performing a model generation task in the model fit dimension is determined from multiple first expert neural network models, and a second target neural network model required for performing a model generation task in the model fitness dimension is determined from multiple first expert neural network models based on the second model weight. The first context flag sequence and the first overall representation of the context are input into the first target neural network model and the second target neural network model, respectively, to obtain a first model prediction result in the model fit dimension and a second model prediction result in the model fitness dimension. The three-dimensional digital twin model is then determined based on the first model prediction result and the second model prediction result. Here, the model fit dimension refers to the degree of fit between the generated 3D digital twin model and the pile foundation column; the model adaptation dimension refers to the degree of adaptation between the generated 3D digital twin model and the performance parameters of the first component and the first simulated service load; furthermore, after obtaining the prediction results of the first model and the second model, the 3D digital twin model can be obtained by weighted summation of the prediction results of the first model and the second model; the weight values ​​used in the weighted summation process can be determined according to actual needs, and this example does not impose any special restrictions on this.

[0062] Secondly, the target remaining life of the three-dimensional digital twin model under extreme service conditions is determined based on the dynamic data of pile foundation load. Specifically, this can be achieved as follows: First, based on the dynamic data of pile foundation load and historical environmental variable data corresponding to historical column status data, the fatigue damage coefficient of the combined effect of waves, tides, and sea winds on the pile foundation column is determined. Second, the critical stress area of ​​the three-dimensional digital twin model is determined, and the original remaining life of the critical stress area under extreme service conditions is determined based on the fatigue damage coefficient. Finally, the cumulative damage of the critical stress area and the crack propagation path caused by the cumulative damage are determined, and the original remaining life is corrected based on the crack propagation path to obtain the target remaining life of the three-dimensional digital twin model under extreme service conditions. In other words, in practical applications, it is first necessary to establish the spatiotemporal correlation between dynamic data of pile foundation loads and historical environmental variable data; then, based on the spatiotemporal correlation, determine the fatigue damage coefficient of the pile foundation column caused by the combined effects of waves, tides, and sea breezes (i.e., the amplification effect of the combined effects of waves, tides, and sea breezes on the pile foundation column); further, calculate the stress distribution of the three-dimensional digital twin model, and determine the critical stress area based on the stress distribution; the specific stress distribution calculation process can be implemented based on three-dimensional simulation software (such as ANSYS Workbench), and this example does not impose any special restrictions on this; at the same time, after obtaining the critical stress... After identifying the stress region, the original remaining life of the critical stress region under extreme service conditions can be determined based on the fatigue damage coefficient using 3D simulation software. Furthermore, to further improve the accuracy of the obtained life, the cumulative damage of the critical stress region can be determined based on the improved Miner criterion, combined with the damage memory effect (such as considering load sequence and overload strengthening / weakening). Then, a phase-field model of material microcrack propagation is introduced, and the crack propagation path caused by the cumulative damage is determined through macro-micro damage coupling by finite element discretization. Finally, the original remaining life is corrected based on the crack propagation path to obtain the target remaining life.

[0063] In step S130, based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data, the neural network model to be trained is trained to obtain a fatigue life prediction model.

[0064] Specifically, the process of determining the fatigue life prediction model can be achieved as follows: A first original feature map is constructed based on the column material parameters, dynamic data of pile load, and historical environmental variable data corresponding to the historical column state data; the first original feature map is input into the neural network model to be trained to obtain the fatigue life prediction result of the pile column; a loss function is constructed based on the fatigue life prediction result and the target remaining life, and the neural network model to be trained is trained based on the loss function to obtain the fatigue life prediction model.

[0065] In one exemplary embodiment, a first original feature map is constructed based on the column material parameters of the pile foundation column, dynamic data of the pile foundation load, and historical environmental variable data corresponding to historical column status data. This includes: using the pile foundation column in an offshore photovoltaic scenario as a first vertex, and the connection relationship between the pile foundation columns as a first connecting edge; determining the first edge weight of the first connecting edge based on the positional relationship between the pile foundation columns, and using the column material parameters of the pile foundation column, dynamic data of the pile foundation load, and historical environmental variable data corresponding to historical column status data as second, third, and fourth vertices; determining the second, third, and fourth connecting edges based on the relationship between the pile foundation column and its material parameters, dynamic data of the pile foundation load, and historical environmental variable data; and constructing the first original feature map based on the first vertex, second vertex, third vertex, fourth vertex, first connecting edge, second connecting edge, third connecting edge, fourth connecting edge, and the first edge weight. The positional relationship described here refers to the distance between two pile foundation columns, which can be determined based on the position coordinates of each pile foundation column. If the distance exceeds a certain threshold, it is determined that there is no connection; otherwise, it is determined that there is a connection between the two pile foundation columns. Specifically, the weight of the first side is inversely proportional to the distance; that is, the closer the distance, the greater the weight of the first side; otherwise, the smaller the weight of the first side. The weights of the second, third, and fourth connecting sides can be set to 1 or other values; this example does not impose any special restrictions on this. Meanwhile, the obtained first original feature map can be referenced... Figure 5 As shown.

[0066] In one example embodiment, the fatigue life prediction result of the pile foundation column is obtained by inputting the first original feature into the neural network model to be trained. This can be achieved as follows: Assume that there are N1 nodes in the first original feature map. M1 edge ;at the same time, Indicates the node in step N The vector (vector dimension is 128); under this premise, the first connecting edge in the first original feature map is updated in the graph convolutional neural network, as shown by the following formulas (1) and (2):

[0067] ;Formula (1)

[0068] ;Formula (2)

[0069] in, It is the Leaky ReLU activation function; Represents a node The set of neighboring nodes, express and The inner product between; These are the parameters of the graph neural network; simultaneously, the attention weights. This represents the strength of the connection between nodes i and k. In this example embodiment, firstly, the parameters of the graph neural network and the initial vectors of each node are randomly initialized. Secondly, the graph convolutional neural network is iterated through the first original feature map. Assuming the maximum value of N is 5 (i.e., including 5 graph convolutional neural networks), the graph neural network obtains the final vector representation of each node (i.e., the pile foundation column) after 5 iterations (i.e., the result of the fifth graph convolution). Finally, after obtaining the result of the fifth graph convolution, it can be input into the classification layer to obtain the fatigue life prediction result. For a specific scenario diagram of the fatigue life prediction process, please refer to... Figure 6 As shown.

[0070] Furthermore, after obtaining the fatigue life prediction results, a loss function can be constructed based on the fatigue life prediction results and the target remaining life. This loss function is then used to train the neural network model to be trained, resulting in a fatigue life prediction model. The loss function described here can be either the mean squared error loss function or the cross-entropy loss function; this example does not impose any special restrictions on either. Furthermore, after obtaining the fatigue life prediction model, Bayesian optimization can be used to dynamically update the model parameters, allowing the fatigue life prediction model to better adapt to environmental changes (such as seasonal fluctuations), thereby improving the accuracy of the obtained life prediction results.

[0071] In step S140, the real-time model state data of the three-dimensional digital twin model is input into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and the real-time fatigue loss of the pile column is determined based on the real-time remaining life probability distribution.

[0072] In this example embodiment, firstly, the real-time remaining lifetime probability distribution of the 3D digital twin model is determined. Specifically, this can be achieved as follows: based on the real-time model state data of the 3D digital twin model and the column material parameters of the 3D digital twin model, a second original feature map is constructed, and node features of the second original feature map are updated based on the first image convolutional layer to obtain the first image convolutional processing result; based on the second image convolutional layer, node features of the first image convolutional processing result are updated to obtain the second image convolutional processing result, and the process of determining the second image convolutional processing result is repeated sequentially to obtain the third image convolutional processing result, ..., the Nth image convolutional processing result; based on the classification layer, the Nth image convolutional processing result is classified to obtain the real-time remaining lifetime probability distribution of the 3D digital twin model. The construction process of the second original feature map is similar to that of the first original feature map, and will not be elaborated further here. Meanwhile, the real-time model state data recorded here can be determined based on the real-time wave buoy data, real-time wind speed and direction data, and real-time pile foundation strain data of the pile foundation column. The column material parameters recorded here are the relevant parameters of the materials used in the pile foundation column (such as elastic modulus, loose ratio, tensile strength, compressive strength, and horizontal bearing capacity), etc. Of course, other parameters can also be included, and this example does not impose any special restrictions on this.

[0073] Furthermore, after obtaining the real-time remaining life probability distribution, the real-time fatigue loss of the pile foundation column can be determined based on the real-time remaining life probability distribution; that is, the real-time damage value can be determined based on the real-time remaining life probability distribution. If the real-time damage value is greater than the critical threshold (e.g., 0.8), an alarm message can be triggered so that maintenance personnel can carry out inspections or use other methods to maintain the pile foundation column based on the alarm message. This example does not impose any special restrictions on this.

[0074] Thus, the method for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems as described in the exemplary embodiments of this disclosure has been fully implemented. Based on the foregoing description, it can be understood that the method for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems as described in the exemplary embodiments of this disclosure can achieve preventative maintenance and reduce unplanned downtime costs; simultaneously, it can predict the damage evolution under extreme loads such as typhoons and storm surges, allowing for early reinforcement measures; furthermore, the method for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems as described in the exemplary embodiments of this disclosure breaks through the linear assumptions and single physical field limitations of traditional fatigue analysis, achieving accurate prediction of fatigue damage to pile foundation columns in offshore photovoltaic scenarios through interdisciplinary integration, providing an innovative method for ensuring the reliability of new marine energy infrastructure.

[0075] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein. For details not disclosed in the apparatus embodiments of this disclosure, please refer to the embodiments of the method disclosed herein.

[0076] This disclosure also provides an example embodiment of a device for determining the fatigue loss of the foundation columns for offshore photovoltaic systems. Specifically, refer to... Figure 7 As shown, the fatigue loss determination device for the pile foundation column of the offshore photovoltaic system may include a pile foundation load dynamic data determination module 710, a target remaining life determination module 720, a fatigue life prediction model determination module 730, and a real-time fatigue loss determination module 740.

[0077] The pile foundation load dynamic data determination module 710 can be used to acquire historical column status data of pile foundation columns in an offshore photovoltaic scenario, and determine the pile foundation load dynamic data of the pile foundation columns based on the historical column status data. The pile foundation load dynamic data is obtained through the following methods: acquiring buoy, anemometer, and displacement sensor data corresponding to the pile foundation columns in the offshore photovoltaic scenario; acquiring historical wave buoy data monitored by the buoy, historical wind speed and direction data monitored by the anemometer, and historical pile foundation strain data monitored by the displacement sensor; determining historical ocean energy spectrum data of the pile foundation columns based on the historical wave buoy data, and determining historical three-dimensional turbulent wind field data of the pile foundation columns based on the historical wind speed and direction data; determining historical fluid-structure interaction data of the pile foundation columns based on the historical pile foundation strain data, and determining historical fluid-structure interaction data of the pile foundation columns based on the historical ocean energy spectrum data and historical three-dimensional turbulent wind field data. The system uses wind field data and historical fluid-structure interaction data to construct dynamic data of the pile foundation column's load. A target remaining life determination module 720 can be used to construct a three-dimensional digital twin model corresponding to the pile foundation column and determine the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load. A fatigue life prediction model determination module 730 can be used to train a neural network model to obtain a fatigue life prediction model based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and historical environmental variable data corresponding to historical column state data. A real-time fatigue loss determination module 740 can be used to input the real-time model state data of the three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model and determine the real-time fatigue loss of the pile foundation column based on the real-time remaining life probability distribution.

[0078] In one exemplary embodiment of this disclosure, a wave spectrum function is determined to determine a spectrum diagram based on the wave spectrum function and historical wave buoy data, and the upward average period of the wave is determined based on the spectrum diagram. Average period of spectral peaks Among them, the upward average period Average period of spectral peaks The following relationship exists between them: ;in, For spectral coefficients; based on the average period of the upper span Average period of spectral peaks and the upward average period Average period of spectral peaks The relationship between the two factors determines the average period. The specific value of the average period is: Based on the upward average period Average period of spectral peaks Average period The height of the waves was used to determine the historical ocean energy spectrum data of the pile foundation columns.

[0079] In one exemplary embodiment of this disclosure, constructing a three-dimensional digital twin model corresponding to the pile foundation column includes: obtaining an original geometric model of the pile foundation column under an offshore photovoltaic scenario, and configuring first component performance parameters for the original geometric model; wherein, the first component performance parameters include at least one of a first elastic modulus, a first Poisson's ratio, a first tensile strength, a first compressive strength, and a first horizontal bearing capacity of the first component material used to prepare the pile foundation column; applying a first simulated service load under simulated service conditions to the original geometric model; wherein, the simulated service conditions are obtained by simulating the actual service conditions of the pile foundation column, and the first simulated service load includes at least one of a first weather scenario load, a first pressure load, and a first support force load borne by the pile foundation column during service; and generating a three-dimensional digital twin model corresponding to the pile foundation column based on the original geometric model, the first component performance parameters, and the first simulated service load.

[0080] In an exemplary embodiment of this disclosure, determining the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load includes: determining the fatigue damage coefficient of the pile foundation column due to the combined effects of waves, tides, and sea winds based on the dynamic data of the pile foundation load and historical environmental variable data corresponding to historical column status data; determining the critical stress area of ​​the three-dimensional digital twin model and determining the original remaining life of the critical stress area under extreme service conditions based on the fatigue damage coefficient; determining the cumulative damage of the critical stress area and the crack propagation path caused by the cumulative damage, and correcting the original remaining life based on the crack propagation path to obtain the target remaining life of the three-dimensional digital twin model under extreme service conditions.

[0081] In one exemplary embodiment of this disclosure, a fatigue life prediction model is obtained by training a neural network model to be trained based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data. This includes: constructing a first original feature map based on the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data; inputting the first original feature map into the neural network model to be trained to obtain the fatigue life prediction result of the pile foundation column; constructing a loss function based on the fatigue life prediction result and the target remaining life, and training the neural network model to be trained based on the loss function to obtain the fatigue life prediction model.

[0082] In one exemplary embodiment of this disclosure, a first original feature map is constructed based on the column material parameters of the pile foundation column, dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column state data. This includes: using the pile foundation column in an offshore photovoltaic scenario as a first vertex, and the connection relationship between the pile foundation columns as a first connecting edge; determining the first edge weight of the first connecting edge based on the positional relationship between the pile foundation columns, and using the column material parameters of the pile foundation column, dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column state data as a second vertex, a third vertex, and a fourth vertex; determining the second connecting edge, a third connecting edge, and a fourth connecting edge based on the relationship between the pile foundation column and the column material parameters, dynamic data of the pile foundation load, and historical environmental variable data; and constructing the first original feature map based on the first vertex, the second vertex, the third vertex, the fourth vertex, the first connecting edge, the second connecting edge, the third connecting edge, the fourth connecting edge, and the first edge weight.

[0083] In an exemplary embodiment of this disclosure, the fatigue life prediction model includes a first graph convolutional layer, a second graph convolutional layer, ..., an Nth graph convolutional layer, and a classification layer. The process of inputting real-time model state data of a three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model includes: constructing a second original feature map based on the real-time model state data of the three-dimensional digital twin model and the column material parameters of the three-dimensional digital twin model; updating the node features of the second original feature map based on the first graph convolutional layer to obtain a first graph convolutional processing result; updating the node features of the first graph convolutional processing result based on the second graph convolutional layer to obtain a second graph convolutional processing result; and repeating the determination process of the second graph convolutional processing result sequentially to obtain a third graph convolutional processing result, ..., an Nth graph convolutional processing result; and classifying the Nth graph convolutional processing result based on the classification layer to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model.

[0084] The specific details of each module in the aforementioned device for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems have been described in detail in the corresponding method for determining the fatigue loss of the pile foundation columns of offshore photovoltaic systems, so they will not be repeated here.

[0085] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0086] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0087] In exemplary embodiments of this disclosure, an electronic device capable of implementing the above-described methods is also provided. Those skilled in the art will understand that various aspects of this disclosure can be implemented as systems, methods, or program products. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: entirely hardware implementations, entirely software implementations (including firmware, microcode, etc.), or implementations combining hardware and software aspects, collectively referred to herein as circuits, modules, or systems.

[0088] The following reference Figure 8 To describe an electronic device 800 according to such an embodiment of the present disclosure. Figure 8 The electronic device 800 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0089] like Figure 8 As shown, the electronic device 800 is manifested in the form of a general-purpose computing device. The components of the electronic device 800 may include, but are not limited to: at least one processing unit 810, at least one storage unit 820, a bus 830 connecting different system components (including storage unit 820 and processing unit 810), and a display unit 840.

[0090] The storage unit stores program code that can be executed by the processing unit 810, causing the processing unit 810 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit 810 can perform actions such as... Figure 1 Step S110: Obtain historical column status data of the pile foundation column in the offshore photovoltaic scenario, and determine the dynamic data of the pile foundation load of the pile foundation column based on the historical column status data; Step S120: Construct a three-dimensional digital twin model corresponding to the pile foundation column, and determine the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load; Step S130: Train the neural network model to be trained based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column status data to obtain a fatigue life prediction model; Step S140: Input the real-time model status data of the three-dimensional digital twin model into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and determine the real-time fatigue loss of the pile foundation column based on the real-time remaining life probability distribution.

[0091] Storage unit 820 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 8201 and / or cache memory 8202, and may further include a read-only memory (ROM) 8203.

[0092] The storage unit 820 may also include a program / utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 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.

[0093] Bus 830 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0094] Electronic device 800 can also communicate with one or more external devices 900 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 800, and / or with any device that enables electronic device 800 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 850. Furthermore, electronic device 800 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 860. As shown, network adapter 860 communicates with other modules of electronic device 800 via bus 830. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0095] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0096] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible implementations, various aspects of this disclosure may also be implemented as a program product including program code that, when run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. A program product for implementing the methods according to embodiments of this disclosure may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of this disclosure is not limited thereto. In this document, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0097] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0098] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0099] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0100] Program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0101] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0102] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention described herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not invented by this disclosure. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

Claims

1. A method for determining the fatigue loss of pile foundation columns for offshore photovoltaic systems, characterized in that, include: Historical state data of the pile foundation columns in an offshore photovoltaic scenario is acquired, and dynamic data of the pile foundation load of the pile foundation columns is determined based on the historical state data. The dynamic data of the pile foundation load is obtained as follows: Buoys, anemometers, and displacement sensors corresponding to the pile foundation columns in the offshore photovoltaic scenario are acquired; historical wave buoy data monitored by the buoys, historical wind speed and direction data monitored by the anemometers, and historical pile foundation strain data monitored by the displacement sensors are acquired; historical ocean energy spectrum data of the pile foundation columns are determined based on the historical wave buoy data; historical three-dimensional turbulent wind field data of the pile foundation columns are determined based on the historical wind speed and direction data; historical fluid-structure interaction data of the pile foundation columns are determined based on the historical pile foundation strain data; and dynamic data of the pile foundation load of the pile foundation columns is constructed based on the historical ocean energy spectrum data, historical three-dimensional turbulent wind field data, and historical fluid-structure interaction data. A three-dimensional digital twin model corresponding to the pile foundation column is constructed, and the target remaining life of the three-dimensional digital twin model under extreme service conditions is determined based on the dynamic data of the pile foundation load. The construction of the three-dimensional digital twin model corresponding to the pile foundation column includes: acquiring the original geometric model of the pile foundation column in an offshore photovoltaic scenario, and configuring first component performance parameters for the original geometric model; wherein the first component performance parameters include at least one of the first elastic modulus, first Poisson's ratio, first tensile strength, first compressive strength, and first horizontal bearing capacity of the first component material used to prepare the pile foundation column; applying a first simulated service load under simulated service conditions to the original geometric model; wherein the simulated service conditions are obtained by simulating the actual service conditions of the pile foundation column, and the first simulated service load includes at least one of the first weather scenario load, first pressure load, and first support force load borne by the pile foundation column during service; and generating a three-dimensional digital twin model corresponding to the pile foundation column based on the original geometric model, the first component performance parameters, and the first simulated service load. Based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data, a fatigue life prediction model is trained on the neural network model to be trained. The training of the neural network model to be trained to obtain the fatigue life prediction model includes: constructing a first original feature map based on the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data; inputting the first original feature map into the neural network model to be trained to obtain the fatigue life prediction result of the pile foundation column; constructing a loss function based on the fatigue life prediction result and the target remaining life, and training the neural network model to be trained based on the loss function to obtain the fatigue life prediction model. The real-time model state data of the three-dimensional digital twin model is input into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and the real-time fatigue loss of the pile column is determined based on the real-time remaining life probability distribution.

2. The method for determining fatigue loss of the pile foundation column of offshore photovoltaic power generation according to claim 1, characterized in that, The historical ocean energy spectrum data of the pile foundation column is determined based on the historical wave buoy data, including: The wave spectrum function is determined, and a spectrum diagram is generated based on the wave spectrum function and historical wave buoy data. The average period of the wave's upward span is then determined based on the spectrum diagram. Average period of spectral peaks Among them, the upward average period Average period of spectral peaks The following relationship exists between them: ;in, These are the spectral coefficients; Based on the average period of the upper cross Average period of spectral peaks and the upward average period Average period of spectral peaks The relationship between the two factors determines the average period. The specific value of the average period is: ; Based on the upward average period Average period of spectral peaks Average period In addition to the wave height, historical ocean energy spectrum data of the pile foundation columns were determined.

3. The method for determining fatigue loss of the pile foundation column of offshore photovoltaic power generation according to claim 1, characterized in that, Determining the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load includes: Based on the dynamic data of the pile foundation load and the historical environmental variable data corresponding to the historical column status data, the fatigue damage coefficient of the pile foundation column caused by the combined effect of waves, tides and sea breeze is determined. The critical stress region of the three-dimensional digital twin model is determined, and the original remaining life of the critical stress region under extreme service conditions is determined based on the fatigue damage coefficient. The cumulative damage in the critical stress region and the crack propagation path caused by the cumulative damage are determined, and the original remaining life is corrected based on the crack propagation path to obtain the target remaining life of the three-dimensional digital twin model under extreme service conditions.

4. The method for determining fatigue loss of the pile foundation column of offshore photovoltaic power generation according to claim 1, characterized in that, A first original feature map is constructed based on the column material parameters, dynamic data of pile foundation load, and historical environmental variable data corresponding to the historical column status data, including: Taking the pile foundation column in the offshore photovoltaic scenario as the first vertex, and the connection relationship between the pile foundation columns as the first connecting edge; The weight of the first side of the first connecting edge is determined based on the positional relationship between the pile foundation columns, and the column material parameters, pile load dynamic data, and historical environmental variable data corresponding to the historical column status data are used as the second, third, and fourth vertices. The second, third, and fourth connecting edges are determined based on the relationship between the pile foundation column and column material parameters, pile foundation load dynamic data, and historical environmental variable data. Based on the first vertex, second vertex, third vertex, fourth vertex, first connecting edge, second connecting edge, third connecting edge, fourth connecting edge, and the weight of the first edge, a first original feature map is constructed.

5. The method for determining fatigue loss of the pile foundation column of offshore photovoltaic power generation according to claim 1, characterized in that, The fatigue life prediction model includes a first graph convolutional layer, a second graph convolutional layer, ..., an Nth graph convolutional layer, and a classification layer; Specifically, the real-time model state data of the three-dimensional digital twin model is input into the fatigue life prediction model to obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, including: Based on the real-time model state data of the three-dimensional digital twin model and the column material parameters of the three-dimensional digital twin model, a second original feature map is constructed, and the node features of the second original feature map are updated based on the convolutional layer of the first map to obtain the convolutional processing result of the first map. The node features of the first graph convolutional processing result are updated based on the second graph convolutional layer to obtain the second graph convolutional processing result. The process of determining the second graph convolutional processing result is repeated sequentially to obtain the third graph convolutional processing result, ..., the Nth graph convolutional processing result. Based on the classification layer, the convolutional processing result of the Nth graph is classified to obtain the real-time remaining lifetime probability distribution of the three-dimensional digital twin model.

6. A device for determining the fatigue loss of pile foundation columns for offshore photovoltaic systems, characterized in that, include: A dynamic data determination module for pile foundation load is used to acquire historical column status data of pile foundation columns in an offshore photovoltaic scenario, and determine the dynamic data of pile foundation load of the pile foundation columns based on the historical column status data. The dynamic data of pile foundation load is obtained through the following methods: acquiring buoy, anemometer, and displacement sensor data corresponding to the pile foundation columns in the offshore photovoltaic scenario; acquiring historical wave buoy data monitored by the buoy, historical wind speed and direction data monitored by the anemometer, and historical pile foundation strain data monitored by the displacement sensor; determining historical ocean energy spectrum data of the pile foundation columns based on the historical wave buoy data; determining historical three-dimensional turbulent wind field data of the pile foundation columns based on the historical wind speed and direction data; determining historical fluid-structure interaction data of the pile foundation columns based on the historical pile foundation strain data; and constructing the dynamic data of pile foundation load of the pile foundation columns based on the historical ocean energy spectrum data, historical three-dimensional turbulent wind field data, and historical fluid-structure interaction data. A target remaining life determination module is used to construct a three-dimensional digital twin model corresponding to the pile foundation column, and to determine the target remaining life of the three-dimensional digital twin model under extreme service conditions based on the dynamic data of the pile foundation load. The construction of the three-dimensional digital twin model corresponding to the pile foundation column includes: acquiring the original geometric model of the pile foundation column in an offshore photovoltaic scenario, and configuring first component performance parameters for the original geometric model; wherein the first component performance parameters include at least one of the first elastic modulus, first Poisson's ratio, first tensile strength, first compressive strength, and first horizontal bearing capacity of the first component material used to prepare the pile foundation column; applying a first simulated service load under simulated service conditions to the original geometric model; wherein the simulated service conditions are obtained by simulating the actual service conditions of the pile foundation column, and the first simulated service load includes at least one of the first weather scenario load, first pressure load, and first support force load borne by the pile foundation column during service; and generating a three-dimensional digital twin model corresponding to the pile foundation column based on the original geometric model, the first component performance parameters, and the first simulated service load. The fatigue life prediction model determination module is used to train a neural network model to obtain a fatigue life prediction model based on the target remaining life, the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and historical environmental variable data corresponding to the historical column state data. The training of the neural network model to obtain the fatigue life prediction model includes: constructing a first original feature map based on the column material parameters of the pile foundation column, the dynamic data of the pile foundation load, and the historical environmental variable data corresponding to the historical column state data; inputting the first original feature map into the neural network model to obtain the fatigue life prediction result of the pile foundation column; constructing a loss function based on the fatigue life prediction result and the target remaining life, and training the neural network model to obtain the fatigue life prediction model based on the loss function. The real-time fatigue loss determination module is used to input the real-time model state data of the three-dimensional digital twin model into the fatigue life prediction model, obtain the real-time remaining life probability distribution of the three-dimensional digital twin model, and determine the real-time fatigue loss of the pile column based on the real-time remaining life probability distribution.

7. 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 method for determining the fatigue loss of the pile foundation column of the offshore photovoltaic system as described in any one of claims 1-5.

8. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the fatigue loss determination method for the pile foundation column of offshore photovoltaic as described in any one of claims 1-5 by executing the executable instructions.

Images