Multi-point mooring cylindrical FPSO dynamic response and mooring cable configuration digital twinning method

Abstract

The invention discloses a multi-point mooring cylindrical FPSO dynamic response and mooring cable configuration digital twinning method. The multi-point mooring cylindrical FPSO dynamic response and mooring cable configuration digital twinning method comprises the following steps that S1, one sea condition combination is taken for analysis; s2, numerical simulation calculation; s3, fitting the configuration of each root mooring line by using a third-order Fourier series; s4, establishing a motion response and database of the multi-point mooring floating type ocean platform; s5, completing the training of a short-term forecasting model and an inversion model; s6, calculating sea conditions which do not appear in the training set through numerical simulation; s7, verifying the accuracy of the model; s8, completing model training and verification corresponding to all combined sea conditions; s9, enabling the model to be matched with the actual situation; and S10, applying the model to an actual platform. According to the multi-point mooring cylindrical FPSO dynamic response and mooring cable configuration digital twinning method, in the using process, data can be continuously updated according to the actual situation, and the accuracy of the model is improved.

Description

Technical Field

[0001] The present invention belongs to the field of ship and ocean engineering technology, and in particular relates to a digital twin method for dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO. Background Art

[0002] Floating FPSOs operate in deepwater for extended periods, potentially facing extreme sea conditions. This can cause significant platform movement and potentially breakage of mooring lines due to exceeding their load limits, posing significant safety risks to personnel and equipment on board. Accurately understanding the FPSO's motion response, mooring line tension, and configuration under various sea conditions can provide early warnings to personnel, enabling them to fully understand the FPSO's motion and respond promptly, effectively ensuring the safe operation of the FPSO and the safety of personnel on board.

[0003] At present, most of the predictions for marine structures at home and abroad focus on the motion response of ships, while there are relatively few predictions for the motion of cylindrical FPSOs. Most predictions are for the motion within three degrees of freedom of the ship, without coupling the platform motion and mooring cable, and without inverting the mooring cable configuration.

[0004] Therefore, it is urgent to design a digital twin method for the dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO to solve the problems mentioned above that the existing technology does not couple the platform motion and mooring cable and does not invert the mooring cable configuration. Summary of the Invention

[0005] In order to solve the technical problems mentioned in the background technology that the existing technology does not predict the coupling of platform motion and mooring cable, and does not invert the mooring cable configuration, a digital twin method for the dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO is provided to solve the above problems.

[0006] To achieve the above objectives, the specific technical solutions of the digital twin method for the dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO of the present invention are as follows:

[0007] A digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO includes the following steps:

[0008] S1. Divide the sea conditions of the sea area according to the wave scatter diagram to obtain multiple sea condition combinations, and select one of the sea condition combinations for analysis;

[0009] S2. Calculating by numerical simulation the six-degree-of-freedom motion response data of the cylindrical FPSO, the mooring line tension data, and the position information of n points on the mooring line under all sea conditions in the combined sea state;

[0010] S3. fitting the configuration of each mooring line by using a third-order Fourier series to obtain a Fourier series coefficient of the configuration of each mooring line;

[0011] S4. Establishing a database of motion response, mooring line tension, and Fourier series coefficients of a multi-point moored floating offshore platform based on the obtained FPSO six-degree-of-freedom motion response data, mooring line tension data, and mooring line configuration Fourier series coefficients;

[0012] S5. Based on the obtained database, the LSTM method is used to complete the training of the platform motion and mooring cable tension short-term prediction model and the mooring configuration inversion model for the selected sea state combination;

[0013] S6. Calculate the sea conditions that do not appear in the training set within the combined sea state range through numerical simulation;

[0014] S7, using the platform dynamic response data calculated in S6 as a test set to test the model to verify the accuracy of the model;

[0015] S8. Repeat the process from S3 to S7 to complete the model training and verification corresponding to all combined sea conditions;

[0016] S9, after obtaining the environmental parameters of the actual operating sea area of ​​the FPSO and the FPSO motion response and mooring line tension data, gradually use the obtained actual data to replace the numerical simulation data in the model to make the model consistent with the actual situation;

[0017] S10. Apply the model to the actual platform to predict the platform motion response and mooring cable tension, and invert the Fourier series fitting coefficients of the mooring cable configuration.

[0018] Furthermore, the spectral peak period selected for the sea condition combination in S1 is 14s to 19s, the significant wave height range is 6m to 9m, and calculations are performed for the wave direction range of 0° to 120° to obtain data. Within this wave direction range, a wave direction is taken at intervals of 15° for calculation.

[0019] Furthermore, in S2, the significant wave height, spectral peak period and wave direction corresponding to each sea condition in the sea condition combination are used as inputs to the hydrodynamic calculation software for numerical simulation to obtain the six-degree-of-freedom motion response data of the multi-point moored floating offshore platform under the sea condition, the mooring cable tension data and the position information of n points on each mooring cable.

[0020] Furthermore, each mooring cable of the cylindrical FPSO in S3 is equipped with a buoy in the middle, and there is a dragging section that has not been lifted. When fitting the configuration of the mooring cable, the buoy and the touchdown point are used as boundaries, and the mooring cable is divided into three sections: the fairlead-buoy section, the buoy-touchdown point section, and the dragging section for fitting.

[0021] Furthermore, the coordinate system for mooring configuration fitting in S3 is defined as taking each anchor chain anchor point as the coordinate system origin, the XOY plane is parallel to the still water surface, the X axis points to the projection of the platform in a static floating state in the XOY plane, and the Z axis is vertically upward. When performing mooring configuration fitting, the following steps are specifically included:

[0022] S31. Using a third-order Fourier series to fit the configuration of the mooring cable projected in the XOY plane, that is, fitting the x and y coordinates of all mooring cable nodes to obtain the configuration of the projected configuration in the XOY plane;

[0023] S32. Integrate the projected length of the mooring line in the XOY plane according to the mooring configuration to obtain an arc length coordinate r, where the origin of the arc length coordinate is also the anchor point;

[0024] S33. Use the third-order Fourier series to perform piecewise fitting on rz to obtain the overall configuration of the mooring cable and the Fourier coefficients of the mooring cable configuration.

[0025] Furthermore, in S5, the maximum and minimum scaling normalization method is used to normalize the training set data, determine the LSTM model parameters and complete the model training. The input of the short-term prediction model of platform motion and mooring cable tension is the ocean environment parameters, the platform motion response and mooring cable tension time history in the past period, and the output is the prediction results of the platform motion response and mooring cable tension time history in the future period; the input of the mooring configuration inversion model is the ocean environment parameters, the platform motion response and mooring cable tension time history over a period of time, and the output is the Fourier series fitting coefficient of the mooring cable configuration in the same time period.

[0026] Furthermore, in S6, sea conditions that do not appear in the sea condition range are selected, and the corresponding significant wave height, spectral peak period and wave direction are used as the input of the hydrodynamic calculation software to perform numerical simulation, thereby obtaining the six-degree-of-freedom motion response data of the multi-point moored cylindrical FPSO under this sea condition, the mooring line tension data and the position information of n points on each mooring line.

[0027] Furthermore, the model is tested using the six-degree-of-freedom motion response data, mooring cable tension data, and mooring cable configuration Fourier coefficients of the multi-point moored cylindrical FPSO obtained in S6 and S7 as a test set. The platform motion response and mooring cable tension in the future are predicted, and the Fourier series corresponding to the mooring cable configuration in the future is inverted. The results are compared with the actual values ​​obtained from the numerical simulation at the corresponding time to verify the accuracy of the model.

[0028] Furthermore, the data monitored by the wave height meter, position and attitude monitoring system and anchor chain monitoring system will be used in S9 to replace the corresponding numerical simulation results in the database to make the database consistent with the actual situation.

[0029] Furthermore, in S10, the sea condition combination in which the platform is located is determined according to the sea condition parameters encountered by the platform, and the corresponding platform dynamic response model is selected to predict the platform motion response and the mooring cable top tension, and to inverse the mooring cable configuration.

[0030] The digital twin method for dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO of the present invention has the following advantages:

[0031] By simultaneously considering the ocean environment parameters, platform motion response, mooring cable top tension and mooring cable configuration Fourier coefficients, the platform six-degree-of-freedom motion response, mooring cable top tension and mooring cable configuration are predicted and inverted. During use, the data can be continuously updated according to actual conditions. During use, the accuracy of the model is improved due to the accumulation of actual data. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A flow chart showing the configuration fitting of the mooring line of the present invention;

[0033] Figure 2 The mooring cable configuration fitting curve in the XOY plane from the buoy to the fairlead hole of the present invention;

[0034] Figure 3 The fitting curve of the mooring cable configuration from the buoy to the fairlead hole section rz of the present invention;

[0035] Figure 4 Comparison between the platform sway prediction value for the next 10 seconds and the actual value Figure 1 ;

[0036] Figure 5 Comparison between the platform sway prediction value for the next 10 seconds and the actual value Figure 2 ;

[0037] Figure 6 Comparison between the platform sway prediction value for the next 10 seconds and the actual value Figure 3 ;

[0038] Figure 7 Comparison between the predicted value of the platform mooring cable 1 tension in the next 10 seconds and the actual value of the present invention;

[0039] Figure 8 Comparison between the predicted value of the platform mooring line 2 tension in the next 10 seconds and the actual value of the present invention;

[0040] Figure 9 The figure shows the comparison between the predicted value of the platform mooring cable 3 tension in the next 10 seconds and the actual value. DETAILED DESCRIPTION

[0041] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] Those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is intended to be within the scope of the present invention and to form different embodiments. For example, in the claims, any one of the claimed embodiments may be used in any combination.

[0043] Please refer to the attached Figure 1 To the attached Figure 9 The present invention describes the digital twin method of the dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO.

[0044] The digital twin method for dynamic response and mooring cable configuration of a multi-point moored cylindrical FPSO in the present invention comprises the following steps:

[0045] S1. Divide the sea conditions of the sea area according to the wave scatter diagram to obtain multiple sea condition combinations, and select one of the sea condition combinations for analysis;

[0046] S2. Calculating by numerical simulation the six-degree-of-freedom motion response data of the cylindrical FPSO, the mooring line tension data, and the position information of n points on the mooring line under all sea conditions in the combined sea state;

[0047] S3. fitting the configuration of each mooring line by using a third-order Fourier series to obtain a Fourier series coefficient of the configuration of each mooring line;

[0048] S4. Establishing a database of motion response, mooring line tension, and Fourier series coefficients of a multi-point moored floating offshore platform based on the obtained FPSO six-degree-of-freedom motion response data, mooring line tension data, and mooring line configuration Fourier series coefficients;

[0049] S5. Based on the obtained database, the LSTM method is used to complete the training of the platform motion and mooring cable tension short-term prediction model and the mooring configuration inversion model for the selected sea state combination;

[0050] S6. Calculate the sea conditions that do not appear in the training set within the combined sea state range through numerical simulation;

[0051] S7, using the platform dynamic response data calculated in S6 as a test set to test the model to verify the accuracy of the model;

[0052] S8. Repeat the process from S3 to S7 to complete the model training and verification corresponding to all combined sea conditions;

[0053] S9, after obtaining the environmental parameters of the actual operating sea area of ​​the FPSO and the FPSO motion response and mooring line tension data, gradually use the obtained actual data to replace the numerical simulation data in the model to make the model consistent with the actual situation;

[0054] S10. Apply the model to the actual platform to predict the platform motion response and mooring cable tension, and invert the Fourier series fitting coefficients of the mooring cable configuration.

[0055] By simultaneously considering the ocean environment parameters, platform motion response, mooring cable top tension and mooring cable configuration Fourier coefficients, the platform six-degree-of-freedom motion response, mooring cable top tension and mooring cable configuration are predicted and inverted. During use, the data can be continuously updated according to actual conditions. During use, the accuracy of the model is improved due to the accumulation of actual data.

[0056] Furthermore, the specific parameters for selecting the sea condition combination in S1 are as follows: a spectral peak period of 14s to 19s, and a significant wave height range of 6m to 9m. Since cylindrical FPSOs are rotating structures and their mooring system uses a three-point mooring, when selecting the wave direction, only the wave direction range of 0° to 120° needs to be calculated to obtain data. Within this wave direction range, wave directions are calculated at intervals of 15°. In this embodiment, the selected sea condition combination is a significant wave height of 6.5 to 8.5m, a spectral peak period of 14.5 to 18.5s, and a wave direction of 90° to 120°.

[0057] Furthermore, in S2, the significant wave height, spectral peak period and wave direction corresponding to each sea condition in the sea condition combination are used as inputs to the hydrodynamic calculation software for numerical simulation to obtain the six-degree-of-freedom motion response data of the multi-point moored floating offshore platform under the sea condition, the mooring cable tension data and the position information of n points on each mooring cable.

[0058] Further, if Figures 1 to 3As shown, each mooring cable of the cylindrical FPSO in S3 is equipped with a buoy in the middle, and there is a part of the dragging section that has not been lifted. When fitting the configuration of the mooring cable, the buoy and the touchdown point are used as boundaries, and the mooring cable is divided into three sections: the fairlead-buoy section, the buoy-touchdown point section and the dragging section for fitting. The coordinate system for fitting the mooring configuration is defined as taking each anchor point of the anchor chain as the origin of the coordinate system, the XOY plane is parallel to the still water surface, the X-axis points to the projection of the platform in a static floating state in the XOY plane, the Z-axis is vertically upward, and the coordinate system satisfies the right-hand rule. When fitting the mooring configuration, the following steps are specifically included:

[0059] S31. Using a third-order Fourier series to fit the configuration of the mooring cable projected in the XOY plane, that is, fitting the x and y coordinates of all mooring cable nodes to obtain the configuration of the projected configuration in the XOY plane;

[0060] S32. Integrate the projected length of the mooring line in the XOY plane according to the mooring configuration to obtain an arc length coordinate r, where the origin of the arc length coordinate is also the anchor point;

[0061] S33. Use the third-order Fourier series to perform piecewise fitting on rz to obtain the overall configuration of the mooring cable and the Fourier coefficients of the mooring cable configuration.

[0062] The constraints in the fitting process are: the fitting curve passes through the fairlead hole, buoy and touchdown point, the mooring line is continuous but not differentiable at the buoy, and is continuous and differentiable at the touchdown point. Figure 1 As shown, the XOY plane of the section from the buoy to the fairlead hole and the fitting curve of the mooring cable configuration in rz are as follows: Figure 2 and Figure 3 shown.

[0063] The Fourier series expression obtained by three-segment fitting is as follows:

[0064] 1) Mopping area:

[0065] The anchor chains in the dragging section are all in a plane parallel to the XOY plane, so only the xy coordinates need to be fitted, and the configuration expression is:

[0066]

[0067] Where y 12 is the y coordinate of the anchor chain at the anchor point, x 12 is the x coordinate of the anchor chain at the anchor point, y 11 is the y coordinate of the anchor chain at the touchdown point, x 11 is the x-coordinate of the anchor chain at the touchdown point.

[0068] Integrate the length of the mooring line in the XOY plane to obtain the arc length coordinate r:

[0069]

[0070] 2) From touchdown point to buoy section:

[0071] The configuration expression of the projection in the XOY plane is:

[0072]

[0073] Where y 22 is the y coordinate of the anchor chain at the buoy, x 22 is the x coordinate of the anchor chain at the buoy, y 21 is the y coordinate of the anchor chain at the touchdown point, x 21 is the x-coordinate of the anchor chain at the touchdown point.

[0074] Integrate the projected length of the mooring cable in the XOY plane to obtain the arc length coordinate r:

[0075]

[0076] Where l1 is the length of the anchor chain from the touchdown point to the anchor point.

[0077] The rz configuration expression is:

[0078]

[0079] Where z 32 is the z coordinate of the anchor chain at the buoy, r 32 is the length of the anchor chain from the buoy to the anchor point, z 31 is the y coordinate of the anchor chain at the touchdown point, r 31 It is the length of the anchor chain from the touchdown point to the anchor point.

[0080] 3) Buoy to fairlead hole section:

[0081] The configuration expression of the projection in the XOY plane is:

[0082]

[0083] Where y 42 is the y coordinate of the anchor chain at the buoy, x 42 is the x coordinate of the anchor chain at the buoy, y 41 is the y coordinate of the anchor chain at the fairlead hole, x 41 is the x-coordinate of the anchor chain at the fairlead.

[0084] Integrate the projected length of the mooring cable in the XOY plane to obtain the arc length coordinate r :

[0085]

[0086] Where l2 is the length of the anchor chain from the buoy to the anchor point.

[0087] The rz configuration expression is:

[0088]

[0089] Where z 52 is the z coordinate of the anchor chain at the buoy, r 52 is the length of the anchor chain from the buoy to the anchor point, z 51 is the y coordinate of the anchor chain at the fairlead hole, r 51 It is the length of the anchor chain from the fairlead to the anchor point.

[0090] The fitting expression of each mooring line configuration contains 41 parameters, and the mooring line configuration can be obtained by inverting these parameters.

[0091] Furthermore, in S5, the platform motion response and mooring cable tension data under each sea condition corresponding to the sea condition combination are extracted from the database as the training set of the LSTM model. The training set data is normalized using the maximum and minimum value scaling normalization method, the LSTM model parameters are determined, and the model training is completed. The input of the short-term prediction model of platform motion and mooring cable tension is the ocean environment parameters, the platform motion response and mooring cable tension time history over the past period of time, and the output is the prediction result of the platform motion response and mooring cable tension time history over the future period of time; the input of the mooring configuration inversion model is the ocean environment parameters, the platform motion response and mooring cable tension time history over a period of time, and the output is the Fourier series fitting coefficient of the mooring cable configuration in the same time period.

[0092] Specifically, the normalization formula is as follows:

[0093]

[0094] in, is the normalized original input value, is the original input value, is the maximum value of the original input value, is the minimum of the original input values.

[0095] Using normalized platform six-degree-of-freedom motion response data and mooring cable tension data as input, and the next 10 seconds of motion response and mooring cable tension time history as output, the model parameters were determined as follows: the network consists of one input layer, one hidden layer, one fully connected layer, and one regression output layer. The hidden layer contains 128 neurons. The model was trained for 200 iterations. The initial learning rate was set to 0.01 and decreased by 50% after 150 iterations to prevent overfitting. The activation functions used were sigmoid and tanh, and the optimizer was Adam. After completing the model training, a prediction model for the platform's dynamic response under this sea condition combination was obtained.

[0096] Furthermore, in S6, sea conditions that do not appear in the sea condition range are selected, and the corresponding significant wave height, spectral peak period and wave direction are used as the input of the hydrodynamic calculation software to perform numerical simulation, thereby obtaining the six-degree-of-freedom motion response data of the multi-point moored cylindrical FPSO under this sea condition, the mooring line tension data and the position information of n points on each mooring line.

[0097] Furthermore, the model is tested using the six-degree-of-freedom motion response data, mooring cable tension data, and mooring cable configuration Fourier coefficients of the multi-point moored cylindrical FPSO obtained in S6 and S7 as a test set. The platform motion response and mooring cable tension in the future are predicted, and the Fourier series corresponding to the mooring cable configuration in the future is inverted. The results are compared with the actual values ​​obtained from the numerical simulation at the corresponding time to verify the accuracy of the model.

[0098] Further, if Figures 4 to 9 As shown, the test set selected in this example has a wave height of 8m, a peak period of 17s, and a wave direction of 110°. This sea condition does not occur in the training set. Based on this sea condition, the platform's sway, roll, pitch, and the tension of three mooring lines are predicted for the next 10 seconds and compared with the actual values ​​obtained from numerical simulation at the corresponding time.

[0099] Error evaluation indicators are introduced to evaluate the forecast accuracy and verify the accuracy of the model.

[0100] In this embodiment, the accuracy of the model is evaluated using three indicators: mean absolute error (MAE), root mean square error (RMSE), and confidence interval of error at a confidence level of 0.95. MAE and RMSE reflect the deviation between the predicted value and the true value of the platform motion response. The confidence interval of error at a confidence level of 0.95 means that there is a 95% probability that the error between the true value and the predicted value will fall within the interval [T1, T2], indicating the upper and lower limits of the prediction error. The closer the predicted value is to the true value, the closer the MAE, RMSE, and the upper and lower limits of the confidence interval T1 and T2 are to 0. The specific calculation formulas are as follows:

[0101]

[0102] Among them, y i is the true value of the platform motion; is the predicted value of platform motion; and s are the mean and standard deviation of the error between the true value and the predicted value respectively; z α2 The Z value corresponding to the confidence level can be found in the Z value table. The Z value at a confidence level of 0.95 is 1.96.

[0103] According to the above formula, the prediction error evaluation table of the platform motion sway, roll, pitch and the tension of three mooring lines of the platform under the test sea conditions is as follows.

[0104]

[0105]

[0106] Table 1 Evaluation table of short-term forecast errors under test sea conditions

[0107] It can be seen from Table 1 that the model has a high prediction accuracy for both motion prediction and mooring tension prediction.

[0108] Furthermore, after the model in S9 is used in an actual FPSO, as the platform operation time increases, the data monitored by the wave height meter, position and attitude monitoring system, and anchor chain monitoring system will replace the corresponding numerical simulation results in the database to make the database consistent with the actual situation.

[0109] Furthermore, in S10, in an actual FPSO, the sea condition combination in which the platform is located is determined according to the sea condition parameters encountered by the platform, and the corresponding platform dynamic response model is selected to predict the platform motion response and the mooring cable top tension, and to inverse the mooring cable configuration.

[0110] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO, characterized by: The following steps are involved: S1. Divide the sea conditions of the sea area according to the wave scatter diagram to obtain multiple sea condition combinations, and select one of the sea condition combinations for analysis; S2. Calculating by numerical simulation the six-degree-of-freedom motion response data of the cylindrical FPSO, the mooring line tension data, and the position information of n points on the mooring line under all sea conditions in the combined sea state; S3. fitting the configuration of each mooring line by using a third-order Fourier series to obtain a Fourier series coefficient of the configuration of each mooring line; S4. Establishing a database of motion response, mooring line tension, and Fourier series coefficients of a multi-point moored floating offshore platform based on the obtained FPSO six-degree-of-freedom motion response data, mooring line tension data, and mooring line configuration Fourier series coefficients; S5. Based on the obtained database, the LSTM method is used to complete the training of the platform motion and mooring cable tension short-term prediction model and the mooring configuration inversion model for the selected sea state combination; S6. Calculate the sea conditions that do not appear in the training set within the combined sea state range through numerical simulation; S7, using the platform dynamic response data calculated in S6 as a test set to test the model to verify the accuracy of the model; S8. Repeat the process from S3 to S7 to complete the model training and verification corresponding to all combined sea conditions; S9, after obtaining the environmental parameters of the actual operating sea area of ​​the FPSO and the FPSO motion response and mooring line tension data, gradually use the obtained actual data to replace the numerical simulation data in the model to make the model consistent with the actual situation; S10. Apply the model to the actual platform to predict the platform motion response and mooring cable tension, and invert the Fourier series fitting coefficients of the mooring cable configuration.

2. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1 is characterized in that: The spectral peak period selected for the sea condition combination in S1 is 14s to 19s, and the significant wave height range is 6m to 9m. Calculations are performed for the wave direction range of 0° to 120° to obtain data. Within this wave direction range, a wave direction is taken at intervals of 15° for calculation.

3. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1 is characterized in that: In S2, the significant wave height, spectral peak period and wave direction corresponding to each sea condition in the sea condition combination are used as inputs to the hydrodynamic calculation software for numerical simulation to obtain the six-degree-of-freedom motion response data of the multi-point moored floating offshore platform under the sea condition, the mooring cable tension data and the position information of n points on each mooring cable.

4. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1 is characterized in that: Each mooring cable of the cylindrical FPSO in S3 is equipped with a buoy in the middle, and there is a dragging section that is not lifted. When fitting the configuration of the mooring cable, the buoy and the touchdown point are used as boundaries, and the mooring cable is divided into three sections: the fairlead-buoy section, the buoy-touchdown point section, and the dragging section for fitting.

5. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 4 is characterized in that: The coordinate system for mooring configuration fitting in S3 is defined as taking each anchor chain anchor point as the coordinate system origin, the XOY plane is parallel to the still water surface, the X axis points to the projection of the platform in a static floating state in the XOY plane, and the Z axis is vertically upward. When performing mooring configuration fitting, the following steps are specifically included: S31. Using a third-order Fourier series to fit the configuration of the mooring cable projected in the XOY plane, that is, fitting the x and y coordinates of all mooring cable nodes to obtain the configuration of the projected configuration in the XOY plane; S32. Integrate the projected length of the mooring line in the XOY plane according to the mooring configuration to obtain an arc length coordinate r, where the origin of the arc length coordinate is also the anchor point; S33. Use the third-order Fourier series to perform piecewise fitting on rz to obtain the overall configuration of the mooring cable and the Fourier coefficients of the mooring cable configuration.

6. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1 is characterized in that: In S5, the maximum and minimum scaling normalization method is used to normalize the training set data, determine the LSTM model parameters and complete the model training. The input of the short-term prediction model of platform motion and mooring cable tension is the ocean environment parameters, the platform motion response and mooring cable tension time history in the past period, and the output is the prediction results of the platform motion response and mooring cable tension time history in the future period; the input of the mooring configuration inversion model is the ocean environment parameters, the platform motion response and mooring cable tension time history over a period of time, and the output is the Fourier series fitting coefficient of the mooring cable configuration in the same time period.

7. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1 is characterized in that: In S6, a sea condition that does not appear in the sea condition range is selected, and its corresponding significant wave height, spectral peak period and wave direction are used as the input of the hydrodynamic calculation software to perform numerical simulation. The six-degree-of-freedom motion response data of the multi-point moored cylindrical FPSO, the mooring line tension data and the position information of n points on each mooring line under this sea condition are obtained.

8. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 7 is characterized in that: The model is tested using the six-degree-of-freedom motion response data, mooring cable tension data, and mooring cable configuration Fourier coefficients of a multi-point moored cylindrical FPSO obtained in S6 and S7 as a test set. The platform motion response and mooring cable tension in the future are predicted, and the Fourier series corresponding to the mooring cable configuration in the future is inverted. The results are then compared with the actual values ​​obtained from numerical simulation at the corresponding time to verify the accuracy of the model.

9. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1, characterized in that: In S9, the data monitored by the wave height meter, position and attitude monitoring system and anchor chain monitoring system will replace the corresponding numerical simulation results in the database to make the database consistent with the actual situation.

10. The digital twin method for dynamic response and mooring line configuration of a multi-point moored cylindrical FPSO according to claim 1, characterized in that: In S10, the sea condition combination in which the platform is located is determined according to the sea condition parameters encountered by the platform, and the corresponding platform dynamic response model is selected to predict the platform motion response and the mooring cable top tension, and to inverse the mooring cable configuration.

Images