Overall construction method of deepwater jacket

Abstract

The invention relates to the technical field of deepwater jacket construction, and provides an overall construction method of a deepwater jacket. The method comprises the following steps: linking a remote sensing database to obtain construction sea area environment data; collecting jacket structure and hoisting equipment data, and constructing multi-dimensional fusion data in combination with environmental data; constructing a digital twinborn platform based on the multi-dimensional fusion data; inputting the multi-dimensional fusion data into a path generation model to generate an initial hoisting path; the paths are finely layered to obtain a layered path set, and the layered path set comprises the steps of lifting, horizontal transfer, positioning adjustment and fixed installation; on the basis of construction of the constraint space, optimizing paths on the digital twin platform, and generating an optimized layered path set; according to the method, the technical problems that in the deepwater jacket construction process, due to marine environment complexity and hoisting equipment performance fluctuation, hoisting path planning is inaccurate, and multiple targets such as safety, efficiency and energy consumption are difficult to balance are solved.

Description

Technical Field

[0001] The present application relates to the field of deep-sea engineering technology, specifically to the field of deep-water jacket construction technology, and especially to a method for overall construction of a deep-water jacket. Background Art

[0002] In the field of deep-sea engineering, the construction of deep-water jackets is an extremely challenging and complex task. This process not only requires extremely high technical precision and safety, but also must face the many uncertainties and complexities brought about by the marine environment. The complexity and uncertainty of the marine environment are important factors that cannot be ignored in the construction of deep-water jackets. Deep-sea areas are often accompanied by extreme climatic conditions, strong currents, complex seabed topography, and unpredictable storms and waves. These environmental factors not only increase the difficulty of construction, but also directly affect the safety and efficiency of lifting operations. For example, strong winds and large waves may cause instability of lifting equipment and increase the risk of operational errors; while the complex and changeable seabed topography requires that the lifting path must be accurate to avoid collisions with seabed obstacles. The performance fluctuations of lifting equipment under different working conditions are also an important factor affecting the planning of lifting paths. The lifting operation of deep-water jackets usually requires the use of heavy equipment such as large floating cranes and lifting belts. These equipment will be affected by many factors when working in deep-sea environments, such as seawater pressure, temperature, salinity, etc. These factors may cause fluctuations in equipment performance, which in turn affects the accuracy and stability of the lifting operation. In addition, the lifting requirements under different working conditions are also different. Changes in factors such as water depth, weight, and size may affect the performance of the lifting equipment. Summary of the invention

[0003] This application provides an overall construction method for a deepwater jacket, aiming to solve the technical problems of inaccurate lifting path planning and difficulty in balancing multiple objectives such as safety, efficiency and energy consumption during the construction of a deepwater jacket due to the complexity of the marine environment and fluctuations in the performance of the lifting equipment.

[0004] In view of the above problems, the present application provides a general construction method of a deepwater jacket.

[0005] The present application provides a method for the overall construction of a deepwater jacket, the method comprising: linking a remote sensing database to obtain environmental data of a target construction sea area; collecting deepwater jacket structure data and lifting equipment performance data, and constructing multidimensional fusion data in combination with the environmental data; constructing a digital twin platform of the target construction sea area based on the multidimensional fusion data, the digital twin platform being used to simulate the construction process of the deepwater jacket in real time; inputting the multidimensional fusion data into a path generation model to generate a plurality of initial lifting paths, wherein the path generation model is constructed based on a deep neural network; performing path refinement layering on the plurality of initial lifting paths to obtain a plurality of layered path sets, the layered path sets comprising a lifting and hoisting stage, a horizontal transfer stage, a positioning and adjustment stage, and a fixed installation stage; based on the construction constraint space, performing multi-objective path optimization on the lifting and hoisting stage, the horizontal transfer stage, the positioning and adjustment stage, and the fixed installation stage on the digital twin platform to generate an optimized layered path set; performing layered path splicing based on the optimized layered path set to obtain an optimized lifting path, inputting the optimized lifting path into the digital twin platform, and executing the construction display of the deepwater jacket.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The above-mentioned overall construction method of deepwater jackets obtains detailed environmental data of the target construction sea area by connecting to the remote sensing database. At the same time, the structural parameters of the deepwater jackets and the performance data of the lifting equipment are collected, and these data are combined with the environmental data to construct a multi-dimensional fusion data set. Subsequently, a digital twin platform is constructed using this multi-dimensional fusion data. This platform can simulate the entire construction process of the deepwater jacket in real time. After that, a path generation model based on a deep neural network is used to generate multiple initial lifting paths from the multi-dimensional fusion data. These paths are preliminary construction plans, and these initial paths are refined and layered to obtain four stages: lifting, horizontal transfer, positioning adjustment, and fixed installation. Each stage has its specific requirements. Then, in the digital twin platform, the paths of each stage are multi-objective optimized according to the construction constraint space. The optimization goal is to balance multiple aspects such as safety, efficiency, and energy consumption to ensure that the lifting process is both safe and efficient. Through optimization, a series of optimized hierarchical path sets are obtained. Finally, these optimized hierarchical paths are spliced ​​to form a complete optimized lifting path, and this lifting path is input into the digital twin platform to perform the construction display of the deepwater jacket. In this way, the optimization effect of the entire lifting process can be seen in the virtual environment, providing strong guidance and support for actual construction. The whole process not only improves construction efficiency, but also reduces construction risks.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of a process flow of a deepwater jacket overall construction method in one embodiment;

[0011] Figure 2 The figure is a flow chart of constructing the constrained space in a method for constructing a deepwater jacket overall in one embodiment. DETAILED DESCRIPTION

[0012] The embodiment of the present application provides an overall method for constructing a deepwater jacket, thereby solving the technical problem that during the construction of the deepwater jacket, the lifting path planning is inaccurate due to the complexity of the marine environment and the performance fluctuations of the lifting equipment, and it is difficult to balance multiple objectives such as safety, efficiency and energy consumption.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0015] Examples, such as Figure 1 As shown, the present application provides a method for overall construction of a deepwater jacket, the method comprising:

[0016] Link to the remote sensing database to obtain environmental data of the target construction sea area.

[0017] In the embodiment of the present application, in order to obtain accurate environmental information of the target construction sea area, the system terminal is linked to the remote sensing database, and various environmental information of the target construction sea area is extracted from the remote sensing database, including current speed, water temperature, salinity, tide, wave height, wind direction and speed, seabed topography, and possible obstacles, etc. These data can not only help the system terminal understand the complexity and uncertainty of the construction sea area, but also provide a basis for subsequent construction planning to ensure the safety and efficiency of the construction process.

[0018] The deepwater jacket structure data and the lifting equipment performance data are collected, and multi-dimensional fusion data is constructed in combination with the environmental data.

[0019] In one embodiment, during the preparation stage for the construction of a deepwater jacket, the system terminal not only needs to understand the environmental conditions of the construction sea area, but also needs to understand the structural characteristics of the jacket itself and the performance parameters of the lifting equipment. This includes collecting detailed structural data such as the size, material, weight distribution, and connection nodes of the jacket, as well as performance data such as the load-bearing capacity, operating radius, and speed control of the lifting equipment. After the collection of these structural data and performance data is completed, the system terminal combines the collected deepwater jacket structural data and lifting equipment performance data with the environmental data obtained from the remote sensing database, and performs multi-dimensional fusion processing to integrate data from different sources and different dimensions to form a comprehensive, three-dimensional, multi-dimensional fusion data. This multi-dimensional fusion data provides data support for subsequent construction planning, which helps to obtain a more reasonable, safe and efficient lifting path.

[0020] Based on the multi-dimensional fusion data, a digital twin platform of the target construction sea area is constructed, and the digital twin platform is used to simulate the construction process of the deepwater jacket in real time.

[0021] In one embodiment, after obtaining the multi-dimensional fusion data, the system terminal inputs the multi-dimensional fusion data into the modeling tool to construct a virtual construction sea area environment model, a virtual jacket model and a virtual lifting equipment model. The virtual construction sea area environment model includes the three-dimensional topography and hydrological characteristics of the construction sea area. At the same time, environmental change factors such as tides and wind speeds are parameterized so that actual construction conditions can be dynamically simulated. This step is to accurately reflect the actual conditions of the construction environment. The virtual jacket model includes the geometric shape of the jacket, the connection method of each component, and the material properties. At the same time, the physical properties of the jacket, such as stiffness and mass distribution, are parameterized to ensure that the actual mechanical properties can be accurately reflected in subsequent simulations. The virtual lifting equipment model includes parameters such as the arm length, lifting capacity, and rotation angle range of the lifting equipment. At the same time, the dynamic characteristics of the equipment, such as the behavior under different loads and operating speeds, are parameterized so as to accurately simulate the operating performance of the equipment in the simulation. Subsequently, the constructed virtual construction sea area environment model, virtual jacket model, and virtual lifting equipment model are packaged and integrated to build a digital twin platform. This digital twin platform can dynamically present the construction process in a virtual environment and make adjustments based on real-time data. After that, the real-time data interface of the digital twin platform is configured to receive data and adjustment parameters from the site, dynamically update the model, and evaluate the adjustment parameters to ensure the synchronization of the virtual environment and real conditions during the construction process, as well as the accuracy of the simulation.

[0022] The multi-dimensional fusion data is input into a path generation model to generate multiple initial lifting paths, wherein the path generation model is constructed based on a deep neural network.

[0023] In one embodiment, in the construction planning of deepwater jackets, the system terminal uses multi-dimensional fusion data as input and transmits it to a path generation model built based on a deep neural network. The design goal of this path generation model is to mine potential lifting paths from complex data. Since the deep neural network has strong learning and nonlinear processing capabilities, it can analyze the inherent connections in the multi-dimensional fusion data, generate multiple initial lifting paths that may meet the requirements, provide the system terminal with a variety of lifting path options, and reduce the uncertainty of the lifting path.

[0024] Furthermore, the present application provides a method for constructing the path generation model, including:

[0025] Historical environmental data, historical deepwater jacket structure data, and historical lifting equipment performance data are obtained to construct historical multi-dimensional fusion data, and training data is constructed by combining historical lifting path data. Based on the training data, a path generation model is constructed through supervised learning and deep neural networks.

[0026] Preferably, in order to construct a model for generating a lifting path, the system terminal collects a large amount of information from multiple historical data sources, including historical environmental data, historical deepwater jacket structure data, and historical lifting equipment performance data, and integrates these data into a multi-dimensional data set, i.e., historical multi-dimensional fusion data. Subsequently, in combination with the historical lifting path data obtained at the same time, these historically planned paths are used as labels, corresponding to each group of data in the historical multi-dimensional fusion data, to construct training data. After that, supervised learning and deep neural networks are used to construct a path generation model. Supervised learning allows the model to learn from known training data, that is, to adjust its own parameters by analyzing historical multi-dimensional fusion data and corresponding historical lifting path data to minimize prediction errors. Taking the long short-term memory network (LSTM) as an example, the system terminal determines the network structure according to data characteristics and business needs, including input layer, LSTM layer, fully connected layer, output layer, etc. Then, the training data is input into the LSTM model, and the input training data is processed step by step, and a hidden state sequence is generated. The output layer of the LSTM model generates a time step output of multiple paths according to the hidden state sequence. These outputs together constitute a path sequence, each of which corresponds to a complete lifting path. The loss value between the path sequence and the historical lifting path data is then calculated by the mean square error. The gradient of the loss function relative to each parameter of the LSTM model is calculated by the back propagation algorithm. The LSTM layer performs back propagation through each time step in the time series to adjust the prediction accuracy of the entire path sequence. The optimization algorithm Adam is used to update the internal weights and biases of the LSTM model according to the calculated gradient, gradually reduce the value of the loss function, and improve the generation accuracy of the path sequence. After each training cycle, the performance of the LSTM model is evaluated using validation data. The evaluation criteria include the generation accuracy and overall error of multiple path sequences, and according to the validation results, hyperparameters such as learning rate, number of LSTM layers, number of hidden units, etc. are adjusted to improve the accuracy of the generated path sequence. When the generation accuracy and overall error tend to be stable, the system terminal stops training and uses test data to evaluate whether the generation accuracy and overall error meet the corresponding threshold requirements to ensure the generalization ability and practicality of the LSTM model on unseen data. When the test passes, the system terminal outputs the current LSTM model as a path generation model. This path generation model can provide a scientific planning scheme for deepwater jacket installation operations based on complex environmental, structural and equipment conditions.

[0027] The multiple initial lifting paths are finely layered to obtain multiple layered path sets, which include a lifting and hoisting stage, a horizontal transfer stage, a positioning and adjustment stage, and a fixed installation stage.

[0028] In one embodiment, after obtaining multiple initial lifting paths, in order to improve the accuracy and efficiency of the lifting operation, the system terminal performs refined layered processing on these initial lifting paths. This process subdivides each initial lifting path into multiple key stages, each of which corresponds to a specific link in the lifting operation, namely the lifting and hoisting stage, the horizontal transfer stage, the positioning and adjustment stage, and the fixed installation stage. Among them, the lifting and hoisting stage is the beginning stage of the lifting operation, focusing on how to smoothly lift the deep-water jacket from the initial position, involving parameters such as lifting weight, lifting speed, lifting height, and rope strength to ensure that the jacket can safely leave the ground or water surface. The horizontal transfer stage is the process of horizontally transferring the jacket to the designated installation position after it is lifted to a certain height, involving transfer paths, transfer speeds, transfer angles, moving directions, etc., to minimize transfer time and costs. The positioning and adjustment stage is the precise positioning adjustment performed when the jacket approaches the installation position, involving fine-tuning length, safety margins, etc., to eliminate minor errors. The fixed installation phase is the fixed installation after the jacket is accurately positioned, which involves connection strength, horizontality, verticality, tightening torque, etc. to ensure the quality and safety of the installation. By finely layering these initial lifting paths, multiple layered path sets can be obtained, each of which contains detailed steps and precautions from lifting to fixed installation. These layered path sets not only help improve the accuracy and efficiency of lifting operations, but also help reduce operational risks and safety hazards.

[0029] Furthermore, the present application provides a method of finely layering the multiple initial hoisting paths to obtain multiple layered path sets, including:

[0030] The multiple initial lifting paths are analyzed in stages, path features and operation nodes are extracted, and multiple path stage identifiers are generated; feature vector decomposition is performed on the initial lifting paths based on the path stage identifiers to generate multiple lifting stages, multiple horizontal transfer stages, multiple positioning adjustment stages, and multiple fixed installation stages; the multiple lifting stages, the multiple horizontal transfer stages, the multiple positioning adjustment stages, and the multiple fixed installation stages are multi-dimensionally integrated to generate the multiple hierarchical path sets.

[0031] Preferably, the system terminal divides each initial hoisting path into several stages according to the operation process. Each stage corresponds to a key operation node in the hoisting operation. The goal of this step is to divide the continuous path data into multiple independent operation stages for further analysis and feature extraction. Then the path data of each stage is time-marked to clarify the position and order of each stage in the entire hoisting path, which is convenient for path feature extraction and operation node identification in subsequent steps. Subsequently, for each segmented operation stage, specific operation steps are identified. For example, in the lifting and hoisting stage, the operation steps include starting to lift, reaching a certain height, stopping lifting, etc. Then extract the key parameters of each operation step, which include but are not limited to the motion state of the equipment (such as lifting speed, rotation angle), load conditions (such as weight distribution), environmental impact (such as wind speed change), etc., to construct multiple operation feature sets. Afterwards, according to the extracted operation parameters, determine the key operation nodes in each stage. These nodes are important conversion points in the operation process, such as nodes from lifting to transportation. Then, based on these key operation nodes and features, generate multiple path stage identifiers. These identifiers are used to describe the start and end of each operation stage, as well as the main operation features of the stage. Then, the system terminal extracts the characteristic parameters of the corresponding stages from multiple operation feature sets according to the determined multiple path stage identifiers, and uses these characteristic parameters to match multiple initial lifting paths, and intercepts the lifting path segments corresponding to each stage. Then, the intercepted paths are filled into the multidimensional vector structure set based on the number of parameters in chronological order to construct the characteristic vectors of each stage. After obtaining the characteristic vectors of each stage, the system terminal performs multidimensional integration on the divided characteristic vectors, that is, integrates the characteristic vectors belonging to the same initial lifting path to form a complete hierarchical path set. In this way, the system terminal obtains multiple hierarchical path sets, each of which contains all the key stages and path information in the entire process from lifting to fixed installation. In this way, the complexity of the lifting path can be better understood and analyzed, providing more accurate and efficient guidance for subsequent lifting operations.

[0032] Based on the construction constraint space, multi-objective path optimization is performed on the digital twin platform for the lifting and hoisting stage, the horizontal transfer stage, the positioning and adjustment stage, and the fixed installation stage to generate an optimized hierarchical path set.

[0033] In one embodiment, the system terminal screens the lifting and hoisting stage, horizontal transfer stage, positioning adjustment stage, and fixed installation stage according to the pre-built construction constraint space, removes the same group of stages that do not meet the construction constraint space, and then inputs the screened lifting and hoisting stage, horizontal transfer stage, positioning adjustment stage, and fixed installation stage into the digital twin platform, and combines the pre-built multi-dimensional performance adjustment function for multiple simulations and optimizations. This process will comprehensively consider multiple goals, such as safety performance, equipment load, time consumption, etc., and continuously adjust the parameters corresponding to each stage to achieve the best comprehensive effect. Finally, the hierarchical paths that meet the adjustment coefficient threshold are sorted into an optimized hierarchical path set. This optimized hierarchical path set can provide scientific guidance for actual construction and improve construction efficiency and quality.

[0034] Further, if Figure 2 As shown, the present application provides a method for constructing the construction constraint space, including:

[0035] The historical multi-dimensional fusion data is segmented in time series to form multi-stage historical data clusters; the multi-stage historical data clusters are multi-dimensionally analyzed to extract and define the operation boundary parameters of each stage to form a stage boundary feature set; based on the stage boundary feature set, a construction constraint space is constructed, and the construction constraint space includes multiple stage subspaces.

[0036] Preferably, the system terminal parses the obtained historical multidimensional fusion data, obtains the timestamp of each data in the historical multidimensional fusion data, and segments the historical multidimensional fusion data according to these timestamps, extracts the data of each stage in the historical multidimensional fusion data, and constructs a multi-stage historical data cluster, each cluster representing a construction stage. Subsequently, multi-dimensional feature extraction is performed on the historical data cluster of each stage, including the operating parameters of the equipment, the changes in environmental conditions, the dynamic response of the jacket structure, etc. Through statistical analysis of the characteristics of each stage, the key parameters affecting the operational safety and stability of each stage are determined. These parameters reflect the operational boundaries between stages, such as load limit, speed threshold, stress upper limit, etc. The key boundary parameters of each stage are then grouped together to form a stage boundary feature set. This feature set defines the operational restrictions and safety boundaries of each stage. Afterwards, based on the obtained stage boundary feature set, the construction process is divided into multiple stage subspaces. Each subspace corresponds to a specific stage, such as lifting and hoisting, horizontal transportation, etc., and its operational restrictions are defined by the boundary parameters of the stage. In each subspace, the system terminal sets the corresponding operation constraints according to the stage boundary feature set. For example, the subspace of the lifting stage includes the upper limit of the crane's load capacity, the lifting speed range, the operating angle limit, etc. Then, the subspaces of each stage are integrated to form an overall construction constraint space. This constraint space ensures that operations at each stage of the construction process are carried out within a safe and effective range, improving the controllability and success rate of the construction.

[0037] Furthermore, the present application provides multi-objective path optimization for the lifting stage, the horizontal transfer stage, the positioning adjustment stage, and the fixed installation stage on the digital twin platform, including:

[0038] Based on the construction constraint space, the lifting and hoisting stage, the horizontal transfer stage, the positioning adjustment stage, and the fixed installation stage of the multiple hierarchical path sets are initially screened to generate an initially screened path set; the initially screened path set is input into the digital twin platform for virtual working condition simulation to obtain a multi-dimensional response data set; multi-dimensional parameter deconstruction and performance mapping are performed on the multi-dimensional response data set to obtain multi-stage performance indicators; multi-objective path optimization is performed on the multiple hierarchical path sets based on the multi-dimensional performance adjustment function and the multi-stage performance indicators to generate an optimized hierarchical path set.

[0039] Optionally, the system terminal loads the operation constraints of each stage from the construction constraint space, which include safety, equipment load, operation speed, environmental conditions, etc. For each stage in the multiple hierarchical path sets, the hierarchical path is matched with the constraints. Specifically, the system terminal checks whether the lifting hierarchical path meets the load capacity, lifting height, speed limit and other conditions in the lifting and lifting stage, checks the balance, equipment bearing capacity and transportation time of the transfer hierarchical path during the horizontal transfer process, checks the fine-tuning length, safety margin and other requirements of the positioning hierarchical path in the positioning adjustment stage, checks the operation stability, equipment coordination and installation accuracy of the installation hierarchical path during the fixed installation process, and adds the hierarchical paths that do not meet the conditions to the abnormal hierarchical paths. Subsequently, the abnormal hierarchical paths are traversed and matched with the hierarchical paths in the multiple hierarchical path sets in the same group, and the hierarchical path sets corresponding to the multiple hierarchical path sets are eliminated, and the hierarchical path sets that meet all the constraints are retained to generate a preliminary screening path set. The path set contains feasible paths for operations at each stage. Afterwards, the system terminal inputs the generated preliminary screening path set into the digital twin platform for virtual simulation. During the simulation process, the key parameters of the digital twin platform are configured according to the hierarchical path set in the initial screening path set, including equipment status, environmental conditions, etc., to truly reproduce the construction process of each path. After the parameter configuration is completed, the digital twin platform gradually executes four stages: lifting, horizontal transfer, positioning adjustment, and fixed installation, and records the performance of each path during the simulation process. After the simulation is completed, the digital twin platform will generate a multi-dimensional response data set, which includes the response parameters of each stage in the path execution process, including actual load, maximum load, number of risk events, path length, power consumption, start time, end time, etc. Then, the system terminal deconstructs the data in the multi-dimensional response data set, decomposes the complex response data into an operational parameter set, and maps it to the performance indicators of each stage according to the deconstructed parameters, forming a complete multi-stage performance indicator set, which includes safety indicators, equipment load, efficiency indicators, energy consumption, and time consumption. Taking the lifting and hoisting stage as an example, the system terminal obtains the safety index by calculating the ratio of the number of risk events in the lifting and hoisting stage to the total number of risk events in the four stages, where risk events include equipment overload, emergency stop, path deviation, etc. The equipment load is obtained by calculating the ratio of the average actual load of the equipment in this stage to the maximum load of the equipment. The efficiency index is obtained by calculating the ratio of the path length to the time consumption in this stage. The energy consumption is obtained by calculating the product of the power consumption and time consumption in this stage. The time consumption is obtained by calculating the difference between the end time and the start time of this stage. After obtaining the multi-stage performance indicator set, the system terminal applies the multi-dimensional performance adjustment function to evaluate and adjust each stage.The multi-dimensional performance adjustment function is used to balance various optimization objectives, ensure the coordination between various objectives, and calculate the adjustment coefficient to iteratively optimize the parameters of each stage, and gradually adjust the operating parameters of each stage to achieve the best under multiple objectives. After multiple rounds of optimization, the system terminal generates an optimized hierarchical path set. This path set performs well in terms of performance indicators at each stage and can meet various needs in actual construction.

[0040] Furthermore, the present application provides a multi-dimensional performance adjustment function, including:

[0041] The multi-dimensional performance adjustment functions are as follows:

[0042] Optionally, the multi-dimensional performance adjustment function generates adjustment coefficients by integrating key indicators of each operation stage to optimize the performance of each stage. This function takes into account multiple factors such as safety, equipment load, efficiency, energy consumption, and time consumption, and adjusts the impact of these factors on the final path optimization through the weight factors of the adjustment coefficients. The multi-dimensional performance adjustment function is as follows: Among them, C i Characterizes the adjustment coefficient of stage i, which is used to adjust the operating parameters of a certain stage in path optimization. i The safety index that characterizes stage i reflects the safety level of the operation in a certain stage. i Characterizes the equipment load at stage i, indicating the load the equipment is subjected to at a certain stage. i Characterizes the efficiency index of stage i and measures the efficiency of the operation in this stage. i Characterizes the energy consumption of stage i, indicating the energy consumption of the equipment at a certain stage. i The time consumption of stage i is the time required to complete a stage. α and β represent the weight factors of the adjustment coefficients, which are used to adjust the influence of different indicators when generating the adjustment coefficients, ensuring that the influence of certain factors can be appropriately highlighted or weakened when all indicators are considered comprehensively. The selection and adjustment of these two factors depend on the specific optimization goals and application scenarios. Through this multi-dimensional performance adjustment function, different factors in each stage can be balanced, so as to ensure that efficient operation can be maintained while ensuring safety, reducing energy consumption and equipment load when optimizing the path.

[0043] Furthermore, the present application provides a method for performing multi-objective path optimization on the multiple hierarchical path sets based on the multi-dimensional performance adjustment function and the multi-stage performance index to generate an optimized hierarchical path set, including:

[0044] The multi-stage performance indicators are parsed to obtain multiple lifting and hoisting stage performance indicators, multiple horizontal transfer stage performance indicators, multiple positioning adjustment stage performance indicators, and multiple fixed installation stage performance indicators; the multiple lifting and hoisting stage performance indicators, multiple horizontal transfer stage performance indicators, multiple positioning adjustment stage performance indicators, and multiple fixed installation stage performance indicators are input into the multidimensional performance adjustment function to generate multiple lifting and hoisting stage adjustment coefficients, multiple horizontal transfer stage adjustment coefficients, multiple positioning adjustment stage adjustment coefficients, and multiple fixed installation stage adjustment coefficients; based on the multiple lifting and hoisting stage adjustment coefficients, the multiple horizontal transfer stage adjustment coefficients, the multiple positioning adjustment stage adjustment coefficients, and the multiple fixed installation stage adjustment coefficients, the multiple hierarchical path sets are parameter adjusted to obtain multiple adjusted hierarchical path sets; path convergence judgment is performed based on the adjustment coefficient threshold to obtain an optimized hierarchical path set.

[0045] Optionally, after obtaining the multi-stage performance index, the system terminal parses the multi-stage performance index, extracts the performance index of each stage such as lifting and hoisting, horizontal transfer, positioning adjustment and fixed installation, and forms multiple lifting and hoisting stage performance indexes, multiple horizontal transfer stage performance indexes, multiple positioning adjustment stage performance indexes, and multiple fixed installation stage performance indexes. The performance indexes of these stages include safety index, equipment load, efficiency index, energy consumption and time consumption. Subsequently, the performance indexes of the multiple stages parsed are respectively input into the multidimensional performance adjustment function. This multidimensional performance adjustment function generates corresponding adjustment coefficients according to the specific conditions of each stage, taking into account factors such as safety, equipment load, efficiency, energy consumption and time consumption, including multiple lifting and hoisting stage adjustment coefficients, multiple horizontal transfer stage adjustment coefficients, multiple positioning adjustment stage adjustment coefficients, and multiple fixed installation stage adjustment coefficients. After that, the generated multiple adjustment coefficients are applied to the hierarchical path set to adjust each parameter in the path set, that is, multiplying the adjustment coefficient with the parameter of the corresponding hierarchical path. After the parameter adjustment, multiple new hierarchical path sets are formed, that is, multiple adjusted hierarchical path sets, to ensure that the performance at each stage is optimized. Then, the system terminal obtains the preset adjustment coefficient threshold as the standard for judging the path convergence. This adjustment coefficient threshold is set based on historical experience and expert advice. If the adjustment coefficients of the four stages of a hierarchical path set are all less than the adjustment coefficient threshold, the hierarchical path set is considered to have converged. At this time, the system terminal outputs the hierarchical path set as an optimized hierarchical path set. The parameters of these optimized hierarchical path sets at each stage have been optimized and meet the set performance requirements. They will be used in subsequent actual construction to ensure the efficiency and safety of construction.

[0046] Furthermore, the present application provides a method for determining path convergence based on an adjustment coefficient threshold to obtain an optimized hierarchical path set, including:

[0047] The multiple lifting and hoisting stage adjustment coefficients, the multiple horizontal transfer stage adjustment coefficients, the multiple positioning adjustment stage adjustment coefficients, and the multiple fixed installation stage adjustment coefficients are homologously matched to obtain multiple stage adjustment coefficient combinations; the multiple stage adjustment coefficient combinations are numerically judged against the adjustment coefficient threshold; for the first stage adjustment coefficient in the stage adjustment coefficient combination that is smaller than the adjustment coefficient threshold, a parameter freezing operation is performed to retain and solidify the adjustment hierarchical path of the first stage adjustment coefficient to generate a frozen hierarchical path; for the second stage adjustment coefficient that is greater than or equal to the adjustment coefficient threshold, a parameter dynamic optimization adjustment is performed to obtain a dynamic hierarchical path; multi-objective path optimization is performed using the dynamic hierarchical path until the adjustment hierarchical paths are all frozen hierarchical paths to obtain an optimized hierarchical path set.

[0048] Optionally, the system terminal classifies and summarizes the adjustment coefficients of different construction stages (lifting, horizontal transfer, positioning adjustment, fixed installation). The adjustment coefficient of each stage represents the optimization requirements and adjustment degree of that stage. Then match the adjustment coefficients of each construction stage in the same hierarchical path set to form a complete stage adjustment coefficient combination. These combinations represent the optimization status of the hierarchical path set at different stages. Subsequently, the adjustment coefficients in each stage adjustment coefficient combination are compared with the adjustment coefficient threshold. The threshold is used to determine whether the parameters of a certain stage still need to be optimized. For the first-stage adjustment coefficient that is less than the adjustment coefficient threshold, a parameter freezing operation is performed, which means that the path parameters of this stage have met the optimization requirements and no further adjustment is required. These frozen stage adjustment coefficients are applied to the corresponding stage, retaining and solidifying these parameters to generate a frozen hierarchical path. The relevant parameters of this path will not be modified in the subsequent optimization process. For the second-stage adjustment coefficient that is greater than or equal to the adjustment coefficient threshold, it means that there is still room for optimization in this stage. The system terminal performs dynamic optimization adjustment on these unfrozen adjustment coefficients, that is, multiplying the stage adjustment coefficients with the parameters of the corresponding construction stage to form a dynamic hierarchical path, and combining it with the frozen hierarchical path as an adjusted hierarchical path set to input into the digital twin platform for simulation, and then repeating the above optimization process for the dynamic hierarchical paths in the adjusted hierarchical path set, gradually adjusting the parameters of each stage to ensure that the path's safety, efficiency, load, energy consumption, and time consumption are optimally balanced. For the remaining hierarchical path sets, the system terminal performs the same operation until all the hierarchical paths in a certain adjusted hierarchical path set are frozen hierarchical paths, and outputs this adjusted hierarchical path set as an optimized hierarchical path set to ensure the safety and efficiency of the construction process.

[0049] Based on the optimized hierarchical path set, hierarchical path splicing is performed to obtain an optimized lifting path, and the optimized lifting path is input into the digital twin platform to execute the construction display of the deepwater jacket.

[0050] In one embodiment, after obtaining the optimized hierarchical path set, the system terminal performs a sequential check on the optimized hierarchical path set to ensure that the order of the stages in the optimized hierarchical path set is correct. For example, the lifting and hoisting stage should be before the horizontal transfer stage, and the fixed installation stage should be after all other stages. Subsequently, the optimized hierarchical paths in the optimized hierarchical path set are spliced ​​in the order of the stages to form a complete optimized lifting path. During the splicing process, attention is paid to the connection points of the paths of each stage to ensure the continuity and smooth transition of the path. If the path connection between the two stages is not smooth or there is a mutation, smoothing is performed, that is, the connection points are averaged to ensure the continuity of the path and the stability of the operation. Afterwards, the generated optimized lifting path is input into the digital twin platform to simulate the real construction process, and the simulation process of the lifting and hoisting stage, the horizontal transfer stage, the positioning adjustment stage, and the fixed installation stage is visualized and digitized for the entire process demonstration, providing reference and guidance for actual construction.

[0051] In summary, the embodiments of the present application have at least the following technical effects:

[0052] The embodiment of the present application obtains environmental data of the construction sea area by linking the remote sensing database, and collects the structural data of the deepwater jacket and the performance data of the lifting equipment at the same time. These data are integrated into multi-dimensional fusion data, and a digital twin platform is constructed based on this data to simulate the construction process of the jacket in real time. Subsequently, the initial lifting path is generated based on the path generation model of the deep neural network, and the layered path set of multiple operation stages is formed through the path refinement and layering. After that, on the digital twin platform, based on the construction constraint space, the path set of each operation stage is multi-objective path optimization to generate an optimized layered path set. On this basis, a complete optimized lifting path is obtained by splicing the layered paths. Finally, the optimized path is input into the digital twin platform for construction display to ensure the executability and optimization effect of the path. These technical effects jointly solve the technical problems of inaccurate lifting path planning and difficulty in balancing multiple objectives such as safety, efficiency and energy consumption during the construction of the deepwater jacket due to the complexity of the marine environment and the fluctuation of the performance of the lifting equipment, and realize the effect of optimizing the lifting path planning, improving the efficiency and safety of the deepwater jacket lifting operation, and ensuring the safe construction of the deepwater jacket and the smooth progress of the overall project.

[0053] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0054] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0055] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for constructing a deepwater jacket, characterized in that: include: Link to remote sensing database to obtain environmental data of target construction sea areas; Collect deepwater jacket structure data and lifting equipment performance data, and build multi-dimensional fusion data in combination with the environmental data; Based on the multi-dimensional fusion data, a digital twin platform of the target construction sea area is constructed, and the digital twin platform is used to simulate the construction process of the deepwater jacket in real time; Inputting the multi-dimensional fusion data into a path generation model to generate a plurality of initial lifting paths, wherein the path generation model is constructed based on a deep neural network; The multiple initial hoisting paths are finely layered to obtain multiple layered path sets, wherein the layered path sets include a lifting stage, a horizontal transfer stage, a positioning adjustment stage, and a fixed installation stage; Based on the construction constraint space, multi-objective path optimization is performed on the digital twin platform for the lifting stage, the horizontal transfer stage, the positioning adjustment stage, and the fixed installation stage to generate an optimized hierarchical path set; Based on the optimized hierarchical path set, hierarchical path splicing is performed to obtain an optimized lifting path, and the optimized lifting path is input into the digital twin platform to execute the construction display of the deepwater jacket.

2. The method for constructing a deepwater jacket as claimed in claim 1, characterized in that: Constructing the construction constraint space includes: Perform time series segmentation on historical multi-dimensional fusion data to form multi-stage historical data clusters; Performing multi-dimensional analysis on the multi-stage historical data clusters, extracting and defining the operation boundary parameters of each stage, and forming a stage boundary feature set; Based on the stage boundary feature set, a construction constraint space is constructed, where the construction constraint space includes a plurality of stage subspaces.

3. The method for constructing a deepwater jacket as claimed in claim 1, characterized in that: The digital twin platform performs multi-objective path optimization on the lifting stage, the horizontal transfer stage, the positioning adjustment stage, and the fixed installation stage, including: Based on the construction constraint space, the lifting and hoisting phase, the horizontal transfer phase, the positioning and adjustment phase, and the fixed installation phase of the multiple hierarchical path sets are preliminarily screened for constraints to generate a preliminarily screened path set; Inputting the preliminary screening path set into the digital twin platform for virtual working condition simulation to obtain a multi-dimensional response data set; Performing multi-dimensional parameter deconstruction and performance mapping on the multi-dimensional response data set to obtain multi-stage performance indicators; Based on the multi-dimensional performance adjustment function and the multi-stage performance indicators, multi-objective path optimization is performed on the multiple hierarchical path sets to generate an optimized hierarchical path set.

4. The method for constructing a deepwater jacket as claimed in claim 3, characterized in that: The multi-dimensional performance adjustment functions are as follows: Among them, C i Characterizes the adjustment coefficient of stage i; S i Characterize the safety index of stage i; L i Characterize the equipment load of stage i; E i Characterizes the efficiency index of stage i; N i Characterizes the energy consumption of stage i; T i It represents the time consumption of stage i; α and β represent the weight factors of the adjustment coefficient, which are used to adjust the influence of each part.

5. The method for constructing a deepwater jacket as claimed in claim 3, characterized in that: Performing multi-objective path optimization on the multiple hierarchical path sets based on the multi-dimensional performance adjustment function and the multi-stage performance indicators to generate an optimized hierarchical path set includes: Parsing the multi-stage performance indicators to obtain multiple lifting stage performance indicators, multiple horizontal transfer stage performance indicators, multiple positioning adjustment stage performance indicators, and multiple fixed installation stage performance indicators; Input the multiple lifting and hoisting stage performance indicators, multiple horizontal transfer stage performance indicators, multiple positioning adjustment stage performance indicators, and multiple fixed installation stage performance indicators into the multidimensional performance adjustment function to generate multiple lifting and hoisting stage adjustment coefficients, multiple horizontal transfer stage adjustment coefficients, multiple positioning adjustment stage adjustment coefficients, and multiple fixed installation stage adjustment coefficients; Based on the multiple lifting stage adjustment coefficients, the multiple horizontal transfer stage adjustment coefficients, the multiple positioning stage adjustment coefficients, and the multiple fixed installation stage adjustment coefficients, the multiple hierarchical path sets are parameter adjusted to obtain multiple adjusted hierarchical path sets; Path convergence judgment is performed based on the adjustment coefficient threshold to obtain an optimized hierarchical path set.

6. The method for constructing a deepwater jacket as claimed in claim 5, characterized in that: Path convergence judgment is performed based on the adjustment coefficient threshold to obtain an optimized hierarchical path set, including: Homologously matching the multiple lifting and hoisting stage adjustment coefficients, the multiple horizontal transfer stage adjustment coefficients, the multiple positioning stage adjustment coefficients, and the multiple fixed installation stage adjustment coefficients to obtain multiple stage adjustment coefficient combinations; Performing numerical judgment on the combination of the multiple stage adjustment coefficients and the adjustment coefficient threshold; For the first stage adjustment coefficient in the stage adjustment coefficient combination that is less than the adjustment coefficient threshold, a parameter freezing operation is performed to retain and solidify the adjustment hierarchical path of the first stage adjustment coefficient to generate a frozen hierarchical path, and for the second stage adjustment coefficient that is greater than or equal to the adjustment coefficient threshold, a parameter dynamic optimization adjustment is performed to obtain a dynamic hierarchical path; The dynamic hierarchical paths are used to perform multi-objective path optimization until all the adjusted hierarchical paths are frozen hierarchical paths, thereby obtaining an optimized hierarchical path set.

7. The method for constructing a deepwater jacket as claimed in claim 6, characterized in that: The multiple initial lifting paths are refined and layered to obtain multiple layered path sets, including: Performing phase-by-phase analysis on the multiple initial lifting paths, extracting path features and operation nodes, and generating multiple path phase identifiers; Based on the path stage identifier, the initial lifting path is decomposed into feature vectors to generate a plurality of lifting stages, a plurality of horizontal transfer stages, a plurality of positioning adjustment stages, and a plurality of fixed installation stages; The multiple lifting stages, the multiple horizontal transfer stages, the multiple positioning adjustment stages, and the multiple fixed installation stages are multi-dimensionally integrated to generate the multiple hierarchical path sets.

8. The method for constructing a deepwater jacket as claimed in claim 1, characterized in that: Constructing the path generation model includes: Obtain historical environmental data, historical deepwater jacket structure data, and historical lifting equipment performance data to build historical multi-dimensional fusion data, and combine historical lifting path data to build training data; Based on the training data, a path generation model is constructed through supervised learning and deep neural network.

Images