Ocean platform dynamic response analysis method and device
By analyzing the dynamic response of marine platforms using a long short-term memory network model, the problem that finite element simulation cannot accurately simulate complex marine environments is solved, and more accurate dynamic response analysis is achieved.
Patent Information
- Application Number
- CN202510872092.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-26
- Publication Date
- 2025-11-21
AI Technical Summary
In existing technologies, finite element simulation cannot accurately simulate complex marine environments, resulting in inaccurate dynamic response analysis results for marine platforms.
A long short-term memory network model is used to collect dynamic analysis data from marine platforms, extract feature parameters, determine the optimal feature subset, and input it into a preset long short-term memory network model to obtain dynamic response data.
This improves the accuracy of dynamic response analysis of marine platforms, enabling it to accurately reflect the real environment of the target marine platform and enhance the reliability of the analysis results.
Smart Images
Figure CN120995062A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of ocean, in particular to a method and device for analyzing dynamic response of an offshore platform. BACKGROUND
[0002] The offshore platform works in the sea area and bears complex dynamic loads for a long time. In order to ensure that the offshore platform can work normally under complex dynamic loads, the staff needs to analyze the dynamic response of the offshore platform to evaluate the health status of the offshore platform.
[0003] At present, the existing technical solution is to analyze the dynamic response by finite element simulation. Specifically, first, a three-dimensional finite element model of the offshore platform is built. Then, the marine environment in which the offshore platform is located is simulated to obtain a simulated environmental load. Finally, the simulated environmental load is applied to the three-dimensional finite element model, and the dynamic analysis module in the finite element analysis software is used to solve the dynamic response of the offshore platform under the simulated environmental load.
[0004] However, for complex marine environments, finite element simulation cannot accurately simulate the real environment, which leads to inaccurate results of the dynamic response analysis of the offshore platform. SUMMARY
[0005] The embodiments of the present application provide a method and device for analyzing the dynamic response of an offshore platform, aiming to solve the technical problem of inaccurate results of the dynamic response analysis of the offshore platform.
[0006] In a first aspect, the embodiments of the present application provide a method for analyzing the dynamic response of an offshore platform, which includes:
[0007] Collecting first dynamic analysis data of a target offshore platform;
[0008] Extracting features from the first dynamic analysis data to obtain a plurality of feature parameters;
[0009] Determining an optimal feature subset from the plurality of feature parameters, the optimal feature subset including at least one feature parameter in the plurality of feature parameters;
[0010] Obtaining a preset long short-term memory network model, the preset long short-term memory network model being used to obtain dynamic response data of the target offshore platform;
[0011] Inputting the optimal feature subset into the preset long short-term memory network model to obtain first dynamic response data of the target offshore platform, the first dynamic response data being used to analyze the dynamic response of the target offshore platform.
[0012] Optionally, the first dynamic analysis data comprises first load condition data of the target offshore platform, the plurality of feature parameters comprises a load condition parameter, and the extracting features from the first dynamic analysis data to obtain a plurality of feature parameters comprises:
[0013] extracting features from the first load condition data to obtain the load condition parameter.
[0014] Optionally, the first dynamic analysis data further comprises inherent characteristic data of the target offshore platform, the plurality of feature parameters further comprises an inherent characteristic parameter, and the extracting features from the first dynamic analysis data to obtain a plurality of feature parameters comprises:
[0015] extracting features from the inherent characteristic data to obtain the inherent characteristic parameter.
[0016] Optionally, the first dynamic analysis data further comprises first environmental load data, the plurality of feature parameters further comprises an environmental load parameter, and the extracting features from the first dynamic analysis data to obtain a plurality of feature parameters comprises:
[0017] extracting features from the first environmental load data to obtain the environmental load parameter.
[0018] Optionally, the obtaining the preset long short-term memory network model comprises:
[0019] constructing a long short-term memory network model;
[0020] collecting a plurality of dynamic analysis data and a plurality of dynamic response data of a target offshore platform, the plurality of dynamic analysis data corresponding one-to-one to the plurality of dynamic response data;
[0021] dividing the plurality of dynamic analysis data and the plurality of dynamic response data into a training set, a test set and a verification set according to a preset allocation ratio;
[0022] using the training set to train the long short-term memory network model by using a K-fold cross-validation method, to obtain parameters of the long short-term memory network model, wherein K is an integer greater than 0;
[0023] based on the parameters of the long short-term memory network model, using the training set, the test set and the verification set to train, test and verify the long short-term memory network model, to obtain optimal parameters of the long short-term memory network model;
[0024] generating the preset long short-term memory network model according to the optimal parameters of the long short-term memory network model and the long short-term memory network model.
[0025] Optionally, the method further comprises:
[0026] collecting second dynamic analysis data of the target offshore platform;
[0027] updating the preset long short-term memory network model according to the second dynamic analysis data.
[0028] Optionally, the second dynamic analysis data comprises second load condition data of the target offshore platform, the second load condition data comprising a cargo weight, and the updating the preset long short-term memory network model according to the second dynamic analysis data comprises:
[0029] determining whether the cargo weight is greater than a first threshold value;
[0030] if yes, updating the preset long short-term memory network model.
[0031] Optionally, the second dynamic analysis data comprises second environmental load data, the second environmental load data comprising a wind speed, and the updating the preset long short-term memory network model according to the second dynamic analysis data comprises:
[0032] determining whether the wind speed is greater than a second threshold value;
[0033] if yes, updating the preset long short-term memory network model.
[0034] Optionally, the updating the preset long short-term memory network model according to the second dynamic analysis data comprises:
[0035] collecting third dynamic analysis data and second dynamic response data of the target offshore platform, the second dynamic response data being dynamic response data measured by the target offshore platform based on the third dynamic analysis data;
[0036] inputting the third dynamic analysis data into the preset long short-term memory network model to obtain third dynamic response data;
[0037] calculating a difference between the second dynamic response data and the third dynamic response data;
[0038] determining whether the difference is greater than a third threshold value;
[0039] if yes, updating the preset long short-term memory network model.
[0040] In a second aspect, the embodiments of the present application further provide an offshore platform dynamic response analysis device, which comprises units for executing the above method.
[0041] In a third aspect, the embodiments of the present application further provide a computer device, which comprises a memory and a processor, the memory has a computer program stored thereon, and the processor implements the above method when executing the computer program.
[0042] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program. The computer program, when executed by a processor, can implement the method described above.
[0043] The embodiments of the present application provide a method and device for analyzing dynamic response of an offshore platform. The method comprises: collecting first dynamic analysis data of a target offshore platform; extracting features from the first dynamic analysis data to obtain a plurality of feature parameters; determining an optimal feature subset from the plurality of feature parameters, the optimal feature subset comprising at least one feature parameter in the plurality of feature parameters; obtaining a preset long short-term memory network model, the preset long short-term memory network model being used to obtain dynamic response data of the target offshore platform; and inputting the optimal feature subset into the preset long short-term memory network model to obtain first dynamic response data of the target offshore platform, the first dynamic response data being used to analyze the dynamic response of the target offshore platform. As can be seen, the embodiments of the present application collect first dynamic analysis data of a target offshore platform and extract features to obtain a plurality of feature parameters. Then, an optimal feature subset is determined from the plurality of feature parameters and input into a preset long short-term memory network model to obtain dynamic response data of the target offshore platform. As can be seen, the first dynamic analysis data is obtained by collection. The feature parameters extracted from the first dynamic analysis data can truly reflect the real environment of the target offshore platform, thereby effectively improving the accuracy of the analysis result of the dynamic response of the target offshore platform. BRIEF DESCRIPTION OF DRAWINGS
[0044] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.
[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.
[0046] One or more embodiments are exemplarily illustrated by pictures in the drawings corresponding thereto, and these exemplary illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, unless otherwise specified. The drawings in the drawings do not constitute a proportional limitation.
[0047] Figure 1a One of the flowcharts of the method for analyzing dynamic response of an offshore platform provided by the embodiments of the present application;
[0048] Figure 1b A structural diagram of a long short-term memory network model provided for an embodiment of the present application;
[0049] Figure 2a A flowchart of a second ocean platform dynamic response analysis method provided for an embodiment of the present application;
[0050] Figure 2b A flowchart of an ocean platform dynamic response model online updating provided for an embodiment of the present application;
[0051] Figure 3 A schematic block diagram of an ocean platform dynamic response analysis device provided for an embodiment of the present application;
[0052] Figure 4 A computer device provided for an embodiment of the present application. DETAILED DESCRIPTION
[0053] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.
[0054] The following disclosure provides many different embodiments, or examples, for implementing different structures of the present application. For the purpose of simplicity, the description of a particular example will not necessarily be repeated in the description of each example. Of course, they are merely examples and are not intended to limit the present application. In addition, reference numerals and / or letters can be repeated in different examples. Such repetition is for the purpose of simplicity and clarity and does not indicate a relationship between the various embodiments and / or arrangements being discussed.
[0055] It should be understood that when used in the specification and the appended claims, the terms "comprise" and "include" indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0056] It should also be understood that the terms used herein in the present application specification are only for the purpose of describing particular embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise.
[0057] It should also be further understood that the term "and / or" as used in the specification and in the claims, if any, means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.
[0058] As used in the specification and in the claims, the term "if' can be interpreted as meaning "when" or "once" or "in response to a determination" or "in response to a detection" depending on the context. Similarly, the phrase "if determined" or "if detected [the described condition or event]" can be interpreted to mean "once determined" or "in response to a determination" or "once detected [the described condition or event]" or "in response to a detection [the described condition or event]" depending on the context.
[0059] In order to solve the technical problem that the dynamic response analysis result of the offshore platform in the prior art is inaccurate, the present application provides an offshore platform dynamic response analysis device, which can improve the accuracy of the dynamic response analysis result of the offshore platform.
[0060] Figure 1a A flowchart of an offshore platform dynamic response analysis method provided by an embodiment of the present application is shown in FIG. 1. In an embodiment, the method comprises:
[0061] S1, collecting first dynamic analysis data of a target offshore platform.
[0062] The first dynamic analysis data is used to analyze the dynamic response of the target offshore platform. The first dynamic analysis data comprises first load condition data, inherent characteristic data and first environmental load data. The first load condition data is the load condition data of the target offshore platform at a first time. The load condition data of the target offshore platform includes but is not limited to weight information of goods, start-stop state information of equipment and anchoring system tension information. The inherent characteristic data is the inherent characteristic data of the target offshore platform at the first time. The inherent characteristic data includes but is not limited to lines section parameter information and structural mass distribution parameter information. The lines section parameter information includes but is not limited to the draft of the target offshore platform, the roll angle of the target offshore platform, the pitch angle of the target offshore platform, the total mass of the target offshore platform and the moment of inertia. The first environmental load data is the environmental load data of the target offshore platform at the first time. The environmental load data includes but is not limited to wind speed, wind direction, wave spectrum density, significant wave height and mean period.
[0063] S2, extracting features from the first dynamic analysis data to obtain a plurality of feature parameters.
[0064] Before extracting the features, the embodiment of the present application performs a preprocessing operation on the collected first dynamic analysis data. The preprocessing operation comprises denoising processing and data completion processing in sequence.
[0065] The embodiments of the present application can use median filtering and moving average filtering to remove random noise in the first dynamic analysis data.
[0066] The embodiments of the present application adopt an interpolation completion method to perform a data completion operation on the missing data in the first dynamic analysis data. Preferably, a cubic spline interpolation method is used to fill in the missing time series data of the first dynamic analysis data.
[0067] In an embodiment, features are extracted from the inherent characteristic data to obtain inherent characteristic parameters. The inherent characteristic parameters include the draft, the angle of roll, the angle of pitch, the total mass, and the moment of inertia.
[0068] In another embodiment, features are extracted from the first load condition data to obtain load condition parameters. The load condition parameters include the weight of the cargo, the start-stop state of the equipment, and the anchor system tension.
[0069] In another embodiment, features are extracted from the first environmental load data to obtain environmental load parameters. The load parameters include the wind speed, the wind direction, the wave spectral density, the significant wave height, and the mean period.
[0070] It should be noted that the inherent characteristic parameters, the load condition parameters, and the load parameters extracted by the embodiments of the present application are mapped to the interval [-1, 1] to eliminate the influence of the scale difference between different features.
[0071] S3, determining an optimal feature subset from the plurality of feature parameters.
[0072] The optimal feature subset includes at least one feature parameter in the plurality of feature parameters.
[0073] S4, obtaining a preset long short-term memory network model.
[0074] The preset long short-term memory network model is used to obtain the dynamic response data of the target offshore platform. The dynamic response data of the target offshore platform includes but is not limited to the roll angle, the pitch angle, and the vertical acceleration. It should be noted that how to obtain the preset long short-term memory network model will be described in detail in the embodiments below. Here, the present application will not be described again.
[0075] S5, inputting the optimal feature subset into the preset long short-term memory network model to obtain first dynamic response data of the target offshore platform.
[0076] The first dynamic response data is the dynamic response data of the target offshore platform. The first dynamic response data is used to analyze the dynamic response of the target offshore platform.
[0077] It should be noted that there is no fixed execution order between S1-S3 and S4. For example, S4 can be executed first and then S1, or S4 can be executed first and then S2. In this regard, the present application does not make any limitation.
[0078] The embodiment of the present application provides a kind of ocean platform dynamic response analysis method.Therein, the method includes: the first dynamic analysis data of target ocean platform is collected;From the first dynamic analysis data, feature is extracted, and a plurality of characteristic parameters are obtained;From the plurality of characteristic parameters, determine the optimal feature subset, the optimal feature subset includes at least one characteristic parameter in the plurality of characteristic parameters;Obtain preset long short-term memory network model, the preset long short-term memory network model is used to obtain the dynamic response data of the target ocean platform;The optimal feature subset is input into the preset long short-term memory network model, and the first dynamic response data of the target ocean platform is obtained, and the first dynamic response data is used to analyze the dynamic response of the target ocean platform.It can be seen from this that the first dynamic analysis data of target ocean platform is collected and feature is extracted, and a plurality of characteristic parameters are obtained.The optimal feature subset is determined from the plurality of characteristic parameters and the optimal feature subset is input into the preset long short-term memory network model, and the dynamic response data of the target ocean platform is obtained.It can be known that the first dynamic analysis data is obtained by collection.The characteristic parameters extracted from the first dynamic analysis data can truly reflect the real environment of target ocean platform, and then effectively improve the accuracy of target ocean platform dynamic response analysis result.
[0079] In an embodiment, the preset long short-term memory network model is obtained, including:
[0080] S41, long short-term memory network model is constructed.
[0081] Please refer to Figure 1b , Figure 1b The structure diagram of long short-term memory network model provided by the embodiment of the present application is provided.The long short-term memory network model includes input layer, hidden layer, full connection layer and output layer.The hidden layer can be composed of multiple long short-term memory units.
[0082] The input parameter of input layer is optimal feature subset.The dimension of optimal feature subset is determined by the number of characteristic parameters included in optimal feature subset.Output layer outputs the dynamic response data of target ocean platform.
[0083] It should be noted that the mean square error is used as the loss function of long short-term memory network model in the embodiment of the present application.In addition, the adaptive moment estimation optimization algorithm is also used to dynamically adjust the learning rate of long short-term memory network model to accelerate the convergence of long short-term memory network model.
[0084] S42, a plurality of dynamic analysis data and a plurality of dynamic response data of target ocean platform are collected.
[0085] The plurality of dynamic analysis data and the plurality of dynamic response data correspond to each other; for example, the plurality of dynamic analysis data includes A dynamic analysis data, and the plurality of dynamic response data includes B dynamic response data. The target offshore platform generates the B dynamic response data based on the A dynamic analysis data.
[0086] S43, divide the plurality of dynamic analysis data and the plurality of dynamic response data into a training set, a test set, and a validation set according to a preset distribution ratio.
[0087] Preferably, the plurality of dynamic analysis data and the plurality of dynamic response data are divided into a training set, a test set, and a validation set according to a ratio of 7:2:1. Each data set includes a plurality of dynamic analysis data and a plurality of dynamic response data, and the plurality of dynamic analysis data and the plurality of dynamic response data in each data set correspond to each other. For example, the training set includes A dynamic analysis data, B dynamic analysis data, C dynamic analysis data, a dynamic response data, b dynamic response data, and c dynamic response data. The A dynamic analysis data corresponds to the a dynamic response data, the B dynamic analysis data corresponds to the b dynamic response data, and the C dynamic analysis data corresponds to the c dynamic response data.
[0088] S44, use the K-fold cross-validation method to train the long short-term memory network model using the training set to obtain parameters of the long short-term memory network model.
[0089] K is an integer greater than 0; preferably, K is 5. The K-fold cross-validation method is used in the embodiment. Specifically, step 1, divide the data in the training set into K non-intersecting subsets. Each subset includes a plurality of dynamic analysis data and a plurality of dynamic response data, and the plurality of dynamic analysis data and the plurality of dynamic response data in each subset correspond to each other. Step 2, select K-1 subsets from the K subsets to train the long short-term memory network model, and the remaining one subset is used to verify the long short-term memory network model. Step 3, calculate the error between the predicted value of the long short-term memory network model and the true value. Step 4, repeat steps 2 and 3 K times to obtain K error values. Step 5, take the average error of the K error values as the performance evaluation index of the long short-term memory network model to determine the parameters of the long short-term memory network model.
[0090] It should be noted that if the error values obtained by continuous n times of calculation do not decrease, the process of training the long short-term memory network model by using the K-fold cross-validation method can be terminated in advance. Wherein, n is less than k, and n is an integer greater than 0. The value of n can be determined according to the value of k. In this application, no limitation is made. Preferably, n is equal to k-2. For example, if the error values obtained by continuous 3 times do not decrease, the process of training the long short-term memory network model by using the K-fold cross-validation method is terminated in advance. In other words, the parameters of the long short-term memory network model can be obtained without repeating 5 times.
[0091] S45, based on the parameters of the long short-term memory network model, training, testing and verifying the long short-term memory network model using the training set, the test set and the verification set, and obtaining the optimal parameters of the long short-term memory network model.
[0092] Specifically, the network search method or the random search method is used to optimize the parameters of the long short-term memory network model using the training set, the test set and the verification set, and obtain the optimal parameters of the long short-term memory network model. Wherein, the optimal parameters include but are not limited to the number of hidden layers, the number of long short-term memory units, the learning rate and the probability of randomly discarding neurons.
[0093] S46, generating a preset long short-term memory network model according to the optimal parameters of the long short-term memory network model and the long short-term memory network model.
[0094] The embodiment of the application generates a preset long short-term memory network model according to the optimal parameters of the long short-term memory network model and the long short-term memory network model.
[0095] Please refer to Figure 2a , Figure 2a A flowchart of a method for analyzing the dynamic response of an offshore platform provided by an embodiment of the application is shown in Figure 2. In an embodiment, the method further comprises:
[0096] S6, collecting second dynamic analysis data of the target offshore platform;
[0097] The second dynamic analysis data is the dynamic analysis data of the target offshore platform at the second time.
[0098] S7, updating the preset long short-term memory network model according to the second dynamic analysis data.
[0099] In an embodiment, the second dynamic analysis data includes second load condition data of the target offshore platform, and the second load condition data includes the weight of the cargo. The preset long short-term memory network model is updated according to the second dynamic analysis data, comprising:
[0100] S71, determining whether the cargo weight is greater than a first threshold value, if yes, performing S72, if no, performing S73.
[0101] The first threshold value is a value greater than 0. The specific value of the first threshold value is set by the applicant according to actual experience. In this application, no limitation is made.
[0102] S72, updating the preset long short-term memory network model.
[0103] Specifically, the embodiment of the application collects a plurality of dynamic analysis data and a plurality of dynamic response data of the target offshore platform in real time. The plurality of dynamic analysis data and the plurality of dynamic response data are divided into a training set, a test set and a validation set, and S45 is repeatedly executed to obtain the optimal parameters of the long short-term memory network model.
[0104] S73, maintaining the preset long short-term memory network model.
[0105] If the cargo weight is less than or equal to the first threshold value, the optimal parameters of the preset long short-term memory network model are not updated.
[0106] In an embodiment, the second dynamic analysis data includes second environmental load data, the second environmental load data includes wind speed, and the updating of the preset long short-term memory network model according to the second dynamic analysis data includes:
[0107] S74, determining whether the wind speed is greater than a second threshold value, if yes, performing S75, if no, performing S76.
[0108] The second threshold value is a value greater than 0. Preferably, the second threshold value is the wind speed value corresponding to a 12-level typhoon. It should be noted that the specific value of the second threshold value can be set by the applicant according to actual experience. In this application, no limitation is made.
[0109] S75, updating the preset long short-term memory network model.
[0110] It should be noted that S75 is the same as or similar to S72. In this application, no further description is made.
[0111] S76, maintaining the preset long short-term memory network model.
[0112] It should be noted that S73 is the same as or similar to S76. In this application, no further description is made.
[0113] In an embodiment, the updating of the preset long short-term memory network model according to the second dynamic analysis data includes:
[0114] S77, collecting third dynamic analysis data and second dynamic response data of the target offshore platform.
[0115] The second dynamic response data is dynamic response data measured by the target offshore platform based on the third dynamic analysis data. The third dynamic analysis data is dynamic analysis data of the target offshore platform at the third time.
[0116] S78, inputting the third dynamic analysis data into a preset long short-term memory network model to obtain third dynamic response data.
[0117] The third dynamic response data is a prediction result of the third dynamic analysis data through the preset long short-term memory network model.
[0118] S79, calculating a difference between the second dynamic response data and the third dynamic response data.
[0119] The second dynamic response data and the third dynamic response data include, but are not limited to, a roll angle, a pitch angle, and a vertical acceleration. Specifically, the embodiment of the present application calculates a difference between the roll angle in the second dynamic response data and the roll angle in the third dynamic response data to obtain a roll angle difference; the embodiment of the present application calculates a difference between the pitch angle in the second dynamic response data and the pitch angle in the third dynamic response data to obtain a pitch angle difference; and the embodiment of the present application calculates a difference between the vertical acceleration in the second dynamic response data and the vertical acceleration in the third dynamic response data to obtain a vertical acceleration difference.
[0120] S710, judging whether the difference is greater than a third threshold value, if yes, performing S711, and if no, performing S712;
[0121] The third threshold value is a value greater than 0. Preferably, the third threshold value is 5%. It should be noted that the specific value of the third threshold value can be set by the applicant according to actual experience. In this regard, the present application does not make any limitation.
[0122] It should be noted that the embodiment of the present application compares the roll angle difference, the pitch angle difference, and the vertical acceleration difference with the third threshold value respectively. If all the differences are greater than the third threshold value, S711 is performed, otherwise, S721 is performed.
[0123] S711, updating the preset long short-term memory network model.
[0124] It should be noted that S711 is the same as or similar to S75 or S72. In this regard, the present application will not be described again.
[0125] S712, keeping the preset long short-term memory network model.
[0126] It should be noted that S712 is the same as or similar to S73 or S76. In this regard, the present application will not be described again.
[0127] It should be noted that please refer to Figure 2b, Figure 2b A flowchart of an online updating process of a dynamic response of an offshore platform is provided in an embodiment of the present application. An embodiment of the present application collects dynamic analysis data of a target offshore platform in real time to update a preset long short-term memory network model, so that the preset long short-term memory network model can adapt to dynamic changes of the target offshore platform, and effectively improve the accuracy of a prediction result of the preset long short-term memory network model.
[0128] Referring to Figure 3 , Figure 3 A schematic block diagram of an offshore platform dynamic response analysis device is provided in an embodiment of the present application. Corresponding to the offshore platform dynamic response analysis method, the present application further provides an offshore platform dynamic response analysis device. The offshore platform dynamic response analysis device includes units for executing the above offshore platform dynamic response analysis method, and the offshore platform dynamic response analysis device can be configured in a terminal such as a desktop computer, a tablet computer, and a laptop computer. Specifically, the offshore platform dynamic response analysis device includes:
[0129] A first collection unit 301 is configured to collect first dynamic analysis data of a target offshore platform.
[0130] An extraction unit 302 is configured to extract features from the first dynamic analysis data to obtain a plurality of feature parameters.
[0131] A determination unit 303 is configured to determine an optimal feature subset from the plurality of feature parameters, and the optimal feature subset includes at least one feature parameter in the plurality of feature parameters.
[0132] A second acquisition unit 304 is configured to acquire a preset long short-term memory network model, and the preset long short-term memory network model is used to acquire dynamic response data of the target offshore platform.
[0133] An input unit 305 is configured to input the optimal feature subset into the preset long short-term memory network model to obtain first dynamic response data of the target offshore platform, and the first dynamic response data is used to analyze the dynamic response of the target offshore platform.
[0134] In an embodiment, the first dynamic analysis data includes first load condition data of the target offshore platform, and the extraction unit 302 is specifically configured to extract features from the first load condition data to obtain the load condition parameters.
[0135] In an embodiment, the first dynamic analysis data includes inherent characteristic data of the target offshore platform, and the extraction unit 302 is further specifically configured to extract features from the inherent characteristic data to obtain the inherent characteristic parameters.
[0136] In an embodiment, the first dynamic analysis data comprises first environmental load data, and the extraction unit 302 is further specifically configured to extract features from the first environmental load data to obtain the environmental load parameter.
[0137] In an embodiment, the second acquisition unit 304 is specifically configured to construct a long short-term memory network model.
[0138] Collect a plurality of dynamic analysis data and a plurality of dynamic response data of a target offshore platform, the plurality of dynamic analysis data and the plurality of dynamic response data correspond one by one;
[0139] According to a preset allocation ratio, the plurality of dynamic analysis data and the plurality of dynamic response data are divided into a training set, a test set and a validation set;
[0140] Using the K-fold cross-validation method, the long short-term memory network model is trained using the training set to obtain parameters of the long short-term memory network model, wherein K is an integer greater than 0;
[0141] Based on the parameters of the long short-term memory network model, the long short-term memory network model is trained, tested and validated using the training set, the test set and the validation set to obtain optimal parameters of the long short-term memory network model;
[0142] According to the optimal parameters of the long short-term memory network model and the long short-term memory network model, the preset long short-term memory network model is generated.
[0143] In an embodiment, the first acquisition unit 301 is further configured to collect second dynamic analysis data of the target offshore platform;
[0144] The device further comprises an updating unit 306, which is configured to update the preset long short-term memory network model according to the second dynamic analysis data.
[0145] In an embodiment, the second dynamic analysis data comprises second load condition data of the target offshore platform, and the second load condition data comprises cargo weight, and the updating unit 306 is further specifically configured to determine whether the cargo weight is greater than a first threshold value;
[0146] If yes, the preset long short-term memory network model is updated.
[0147] In an embodiment, the second dynamic analysis data comprises second environmental load data, and the second environmental load data comprises wind speed, and the updating unit 306 is further specifically configured to determine whether the wind speed is greater than a second threshold value;
[0148] If yes, the preset long short-term memory network model is updated.
[0149] In an implementation, the updating unit 306 further specifically collects third dynamic analysis data of the target offshore platform and second dynamic response data of the target offshore platform, the second dynamic response data being dynamic response data measured by the target offshore platform based on the third dynamic analysis data;
[0150] inputting the third dynamic analysis data into the preset long short-term memory network model to obtain third dynamic response data;
[0151] calculating a difference between the second dynamic response data and the third dynamic response data;
[0152] determining whether the difference is greater than a third threshold value;
[0153] If yes, updating the preset long short-term memory network model.
[0154] As shown in Figure 4 The embodiment of the present application provides a computer device, which comprises a processor 41, a communication interface 42, a memory 43 and a communication bus 44, wherein the processor 41, the communication interface 42 and the memory 43 complete mutual communication through the communication bus 44, and the memory 43 is used for storing a computer program.
[0155] In an embodiment of the present application, the processor 41 is used for executing the program stored in the memory 43, so as to realize the control method of the offshore platform dynamic response analysis provided by any one of the foregoing method embodiments.
[0156] Those skilled in the art can understand that all or part of the processes in the method of the foregoing embodiment can be completed by a computer program instructing related hardware. The computer program can be stored in a storage medium, which is a computer readable storage medium. The computer program is executed by at least one processor in the computer system, so as to realize the process steps of the foregoing method embodiment.
[0157] Therefore, the embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the offshore platform dynamic response analysis method provided by any one of the foregoing method embodiments.
[0158] The storage medium is an entity, non-transient storage medium, for example, can be a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk or an optical disk and various entity storage media that can store program codes. The computer readable storage medium can be non-volatile or volatile.
[0159] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in a general manner. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.
[0160] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of each unit is only a logical functional division, and actual implementation can have another division manner. For example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed.
[0161] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the device embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.
[0162] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application.
[0163] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.
[0164] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, these modifications and variations of the present application are intended to be included within the scope of the claims of the present application and their equivalents.
[0165] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. A method for analyzing the dynamic response of an offshore platform, characterized in that, include: Collect first dynamic analysis data of the target marine platform, which is used to analyze the dynamic response of the target marine platform; Features are extracted from the first dynamic analysis data to obtain multiple feature parameters; Determine an optimal feature subset from the plurality of feature parameters, wherein the optimal feature subset includes at least one feature parameter from the plurality of feature parameters; A preset long short-term memory network model is obtained, which is used to acquire the dynamic response data of the target marine platform; The optimal feature subset is input into the preset long short-term memory network model to obtain the first dynamic response data of the target marine platform. The first dynamic response data is used to analyze the dynamic response of the target marine platform.
2. The method according to claim 1, characterized in that, The first dynamic analysis data includes the first loading condition data of the target marine platform, and the plurality of feature parameters include loading condition parameters. The extraction of features from the first dynamic analysis data to obtain the plurality of feature parameters includes: Features are extracted from the first load data to obtain the load parameters.
3. The method according to claim 2, characterized in that, The first dynamic analysis data also includes inherent characteristic data of the target marine platform, and the plurality of characteristic parameters also include inherent characteristic parameters. The extraction of features from the first dynamic analysis data to obtain the plurality of characteristic parameters includes: Features are extracted from the inherent feature data to obtain the inherent feature parameters.
4. The method according to claim 3, characterized in that, The first dynamic analysis data also includes first environmental load data, and the plurality of feature parameters also include environmental load parameters. The step of extracting features from the first dynamic analysis data to obtain the plurality of feature parameters includes: Features are extracted from the first environmental load data to obtain the environmental load parameters.
5. The method according to claim 4, characterized in that, The acquisition of the preset long short-term memory network model includes: Construct a long short-term memory network model; Multiple dynamic analysis data and multiple dynamic response data of the target marine platform are collected, and the multiple dynamic analysis data and the multiple dynamic response data correspond one-to-one; The plurality of dynamic analysis data and the plurality of dynamic response data are divided into training set, test set and validation set according to a preset allocation ratio; The long short-term memory network model is trained using the training set by employing K-fold cross-validation to obtain the parameters of the long short-term memory network model, where K is an integer greater than 0; Based on the parameters of the Long Short-Term Memory (LSTM) network model, the LTM network model is trained, tested, and validated using the training set, the test set, and the validation set to obtain the optimal parameters of the LTM network model. The preset long short-term memory network model is generated based on the optimal parameters of the long short-term memory network model and the long short-term memory network model itself.
6. The method according to claim 5, characterized in that, The method further includes: Collect the second dynamic analysis data of the target marine platform; The preset long short-term memory network model is updated based on the second dynamic analysis data.
7. The method according to claim 6, characterized in that, The second dynamic analysis data includes the second cargo condition data of the target marine platform, which includes cargo weight. Updating the preset long short-term memory network model based on the second dynamic analysis data includes: Determine whether the weight of the goods exceeds a first threshold; If so, update the preset long short-term memory network model.
8. The method according to claim 6, characterized in that, The second dynamic analysis data includes second environmental load data, which includes wind speed. Updating the preset long short-term memory network model based on the second dynamic analysis data includes: Determine whether the wind speed is greater than the second threshold. If so, update the preset long short-term memory network model.
9. The method according to claim 6, characterized in that, The step of updating the preset long short-term memory network model based on the second dynamic analysis data includes: Collect third dynamic analysis data and second dynamic response data of the target marine platform, wherein the second dynamic response data is the dynamic response data of the target marine platform measured based on the third dynamic analysis data; The third dynamic analysis data is input into the preset long short-term memory network model to obtain the third dynamic response data; Calculate the difference between the second dynamic response data and the third dynamic response data; Determine whether the difference is greater than the third threshold; If so, update the preset long short-term memory network model.
10. A dynamic response analysis device for marine platforms, characterized in that, include: The first acquisition unit is used to acquire the first dynamic analysis data of the target marine platform; An extraction unit is used to extract features from the first dynamic analysis data to obtain multiple feature parameters; A determining unit is configured to determine an optimal feature subset from the plurality of feature parameters, wherein the optimal feature subset includes at least one feature parameter among the plurality of feature parameters; The second acquisition unit is used to acquire a preset long short-term memory network model, which is used to acquire the dynamic response data of the target marine platform; The input unit is used to input the optimal feature subset into the preset long short-term memory network model to obtain the first dynamic response data of the target marine platform. The first dynamic response data is used to analyze the dynamic response of the target marine platform.