An adaptive seabed load optimization method and apparatus

Abstract

This invention provides an adaptive seabed load optimization method and apparatus, relating to the field of marine engineering technology. The method includes: acquiring physical parameters of seabed soil samples, load distribution of marine engineering equipment, and mechanical response data during the interaction between the marine engineering equipment and the seabed; constructing a seabed mechanics simulation model; training a reinforcement learning model based on the seabed mechanics simulation model and the simulation result dataset to generate an optimal load distribution scheme; predicting the seabed response results under different load conditions using the trained neural network; and dynamically adjusting the load distribution of the marine engineering equipment according to the seabed response results and the optimal load distribution scheme. This invention enables real-time adaptive optimization of the load distribution of marine engineering equipment, not only improving the stability and efficiency of marine engineering equipment operation but also significantly reducing seabed deformation and minimizing environmental impact.

Description

Technical Field

[0001] This invention relates to the field of marine engineering technology, and specifically to an adaptive seabed load optimization method and apparatus. Background Technology

[0002] Marine engineering equipment such as underwater robots, tracked equipment, and sleds are widely used in marine resource extraction and maintenance tasks. Their load distribution directly affects the stress and deformation characteristics of the seabed, which is related to the stability of equipment operation and the protection of the seabed environment.

[0003] However, existing technologies still have shortcomings in optimizing load distribution and are unable to cope with the complex and ever-changing seabed environment. Current seabed mechanical analyses mostly employ static or linear models, which, while suitable for simple scenarios, lack sufficient accuracy to address the diversity of seabed soil and the dynamic operation of equipment. For example, the stress distribution and deformation characteristics of soft and hard soils differ significantly, and traditional models struggle to reflect the nonlinear interaction between equipment and the seabed. Furthermore, most optimization methods are based on fixed parameter settings, lacking real-time adjustment capabilities and failing to adapt to changes in soil parameters and environmental conditions during operation, leading to decreased equipment operating efficiency and potential damage to the seabed environment. On the other hand, seabed load optimization involves complex mechanical calculations. In multivariate coupling and dynamic load scenarios, traditional methods suffer from low computational efficiency and poor real-time performance, failing to quickly respond to actual needs.

[0004] Therefore, there is an urgent need for an adaptive optimization method that combines big data, reinforcement learning, and neural networks. Through dynamic modeling, real-time optimization, and efficient feedback mechanisms, it can achieve intelligent adjustment of load distribution, thereby improving equipment operating efficiency, reducing seabed deformation, and providing reliable technical support for complex marine environments. Summary of the Invention

[0005] The purpose of this invention is to provide an adaptive seabed load optimization method and apparatus to address the problems of existing seabed optimization methods failing to reflect the nonlinear interaction between equipment and the seabed; lacking real-time adjustment capabilities, leading to decreased equipment operating efficiency and potential damage to the seabed environment; and traditional methods having low computational efficiency, poor real-time performance, and inability to quickly respond to actual needs. This invention enables intelligent adjustment of load distribution, reflects the nonlinear interaction between equipment and the seabed, thereby improving equipment operating efficiency, reducing seabed deformation, and providing reliable technical support for complex marine environments.

[0006] To achieve the above objectives, in a first aspect, the present invention provides an adaptive seabed load optimization method, comprising:

[0007] Step 1: Obtain the physical parameters of seabed soil samples, the load distribution of marine engineering equipment, and the mechanical response data during the interaction between marine engineering equipment and the seabed;

[0008] Step 2: Using physical parameters and load distribution of marine engineering equipment as inputs and mechanical response data as outputs, construct a seabed mechanical simulation model using simulation software to simulate the interaction between marine engineering equipment and the seabed under different load conditions, and obtain the simulation result dataset.

[0009] Step 3: Define the basic elements of the preset reinforcement learning model. Based on the seabed mechanics simulation model and simulation result dataset, construct a virtual training environment to simulate the operation of marine engineering equipment under different load distribution strategies, train the reinforcement learning model, and optimize the load distribution strategy through multiple interactions and iterations with the virtual training environment to generate the optimal load distribution scheme.

[0010] Step 4: Using physical parameters and load distribution of marine engineering equipment as input parameters and mechanical response data as output parameters, construct a neural network, train the neural network using the backpropagation algorithm, and predict the seabed response results under different load conditions through the trained neural network.

[0011] Step 5: Based on the seabed response results and the optimal load distribution scheme, dynamically adjust the load distribution of marine engineering equipment.

[0012] According to the adaptive seabed load optimization method provided by the present invention, the physical parameters include soil density, internal friction angle, cohesion, and compression modulus, and the mechanical response data include seabed stress distribution, contact pressure distribution, and seabed surface deformation.

[0013] According to an adaptive seabed load optimization method provided by the present invention, the marine engineering equipment includes a seabed robot, tracked equipment, and a sled.

[0014] According to the adaptive seabed load optimization method provided by the present invention, step 1 further includes: cleaning the physical parameters and mechanical response data, removing noise and outliers, extracting key variables and performing standardization processing, and then storing them in a data lake and data warehouse, and using a distributed storage system to realize data storage and management.

[0015] According to the adaptive seabed load optimization method provided by the present invention, in step 2, a seabed mechanical simulation model is constructed using the discrete element method or the finite element method, and the simulation software is EDEM or ABAQUS.

[0016] According to the adaptive seabed load optimization method provided by the present invention, in step 3, the basic elements include state space, action space and reward function. The state space is the current load distribution of marine engineering equipment, seabed stress distribution, contact pressure distribution and seabed surface deformation. The action space is the strategy for adjusting the load distribution. The reward function is to minimize the seabed surface deformation and the energy consumption of marine engineering equipment.

[0017] According to the adaptive seabed load optimization method provided by the present invention, in step 3, a reinforcement learning model is trained using a deep Q-learning algorithm or a policy gradient algorithm.

[0018] According to the adaptive seabed load optimization method provided by the present invention, the neural network includes a multilayer feedforward neural network or a convolutional neural network.

[0019] According to the adaptive seabed load optimization method provided by the present invention, the prediction accuracy of the neural network is controlled within 5%.

[0020] In a second aspect, the present invention provides an adaptive seabed load optimization device, comprising:

[0021] The data acquisition module is used to acquire physical parameters of seabed soil samples, load distribution of marine engineering equipment, and mechanical response data during the interaction between marine engineering equipment and the seabed.

[0022] The mechanical simulation module is used to construct a seabed mechanical simulation model by taking physical parameters and load distribution of marine engineering equipment as inputs and mechanical response data as outputs. It simulates the interaction between marine engineering equipment and the seabed under different load conditions and obtains a simulation result dataset.

[0023] The reinforcement learning optimization module is used to define the basic elements of the preset reinforcement learning model. Based on the seabed mechanics simulation model and simulation result dataset, it constructs a virtual training environment to simulate the operation of marine engineering equipment under different load distribution strategies, trains the reinforcement learning model, and optimizes the load distribution strategy through multiple interactions and iterations with the virtual training environment to generate the optimal load distribution scheme.

[0024] The modeling and prediction module is used to construct a neural network with physical parameters and load distribution of marine engineering equipment as input parameters and mechanical response data as output parameters. The neural network is trained using the backpropagation algorithm and the trained neural network is used to predict the seabed response under different load conditions.

[0025] The adaptive module is used to dynamically adjust the load distribution of marine engineering equipment based on the seabed response results and the optimal load distribution scheme.

[0026] The technical solution of the present invention has at least the following technical effects:

[0027] This invention provides an adaptive seabed load optimization method and apparatus. The method includes: acquiring physical parameters of seabed soil samples, load distribution of marine engineering equipment, and mechanical response data during the interaction between marine engineering equipment and the seabed; constructing a seabed mechanical simulation model; training a reinforcement learning model based on the seabed mechanical simulation model and simulation result dataset to generate an optimal load distribution scheme; predicting the seabed response under different load conditions using a trained neural network; and dynamically adjusting the load distribution of the marine engineering equipment according to the seabed response results and the optimal load distribution scheme. This invention achieves high-precision seabed load optimization and real-time dynamic adjustment through the organic combination of data acquisition and preprocessing, mechanical simulation modeling, reinforcement learning optimization, and neural network prediction. Compared with traditional technologies, it features high modeling accuracy, strong real-time performance, and strong adaptability, effectively meeting the optimization needs in complex seabed environments and providing strong technical support for the design and operation of marine engineering equipment. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0029] In the attached diagram:

[0030] Figure 1 This is a flowchart of the adaptive seabed load optimization method of the present invention;

[0031] Figure 2 This is a structural block diagram of the adaptive seabed load optimization device of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0033] The following detailed description of some embodiments of the present invention will be provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0034] Please see Figure 1This invention provides an adaptive seabed load optimization method based on big data, reinforcement learning, and neural networks. By combining high-precision mechanical modeling and intelligent optimization strategies, it achieves real-time adaptive optimization of load distribution for marine engineering equipment. This not only improves the stability and efficiency of equipment operation but also significantly reduces seabed deformation and environmental impact. This invention is applicable to complex and variable seabed environments, particularly in scenarios where marine engineering equipment operates under complex conditions and load optimization requirements change dynamically. The method includes:

[0035] Step 1: Obtain the physical parameters of seabed soil samples, the load distribution of marine engineering equipment, and the mechanical response data during the interaction between marine engineering equipment and the seabed;

[0036] Specifically, the physical parameters of seabed soil samples include soil density, internal friction angle, cohesion, and compression modulus, used to describe the mechanical properties of the soil under different seabed conditions. Mechanical response data during the interaction between marine engineering equipment and the seabed include seabed stress distribution, contact pressure distribution, and seabed surface deformation, used to characterize the mechanical relationship between the marine engineering equipment and the seabed. Marine engineering equipment includes underwater robots, tracked vehicles, and sleds, among others.

[0037] In some embodiments, step 1 involves collecting soil samples from different seabed areas using sampling equipment, recording the geographical location and basic characteristics of the soil samples, thereby obtaining the physical parameters of the seabed soil samples. Subsequently, the sampling equipment utilizes pressure sensors, displacement sensors, and strain sensors to collect mechanical response data in real time during the interaction between marine engineering equipment and the seabed. To ensure data quality, the collected raw data needs to be preprocessed; that is, step 1 also includes cleaning the physical parameters and mechanical response data, removing noise and outliers, to ensure the accuracy and reliability of the data. Next, key variables, such as stress distribution, load changes, and seabed surface deformation, are extracted from the cleaned data using feature extraction techniques for subsequent analysis and modeling. Then, data standardization is performed to convert the extracted feature data into a standardized format and range, eliminating the influence of different dimensions on model training. The standardized data is then transferred to the storage management module and stored in a data lake and data warehouse, utilizing distributed storage systems (such as Hadoop HDFS and Amazon S3) to achieve efficient data storage and management. Finally, all preprocessed data are integrated to generate a high-quality dataset in a unified format, providing data support for subsequent mechanics simulations, reinforcement learning optimization, and neural network predictions. The entire process fully utilizes big data processing tools (such as Apache Spark) and distributed storage technology to ensure efficient and scalable data processing.

[0038] It's important to note that data lakes can store raw and semi-structured data, supporting diverse data types and formats; data warehouses can store structured and processed data, optimizing query performance and data analysis efficiency; and distributed storage systems provide high availability and fault tolerance, supporting linear data expansion. Through ETL (Extract, Transform, Load) processes, and utilizing big data processing tools such as Apache Spark, data is extracted from the data lake, transformed, and loaded into the data warehouse for data cleaning, transformation, and aggregation. This provides data access interfaces, including APIs and query tools, for subsequent application modules (such as seabed mechanics simulation models, reinforcement learning model optimization, and neural network construction and prediction), enabling rapid data acquisition and sharing. The entire big data storage and management architecture, through efficient data flow and distributed computing technologies, ensures centralized data management, rapid access, and efficient processing, fully leveraging the crucial role of big data technology in the storage management module.

[0039] Step 2: Using physical parameters and load distribution of marine engineering equipment as inputs and mechanical response data as outputs, construct a seabed mechanical simulation model using simulation software to simulate the interaction between marine engineering equipment and the seabed under different load conditions, and obtain the simulation result dataset.

[0040] Specifically, simulation software (such as EDEM and ABAQUS) can be used to construct a seabed mechanics simulation model based on the discrete element method (DEM) or the finite element method (FEM) to simulate the interaction between marine engineering equipment and the seabed under different load conditions, obtain a simulation result dataset, and store the simulation result dataset in the storage management module.

[0041] In some embodiments, a suitable simulation method, such as the discrete element method or the finite element method, is first selected based on the seabed soil and equipment characteristics. Then, the inputs to the seabed mechanics simulation model are set, including soil density, internal friction angle, cohesion, compression modulus, and load distribution of marine engineering equipment (such as the load distribution of tracks and sleds). Next, simulation software (such as EDEM or ABAQUS) is used to construct the seabed mechanics simulation model to simulate the interaction between the marine engineering equipment and the seabed. After the seabed mechanics simulation model is completed, it is run to simulate key mechanical responses such as seabed stress distribution, contact pressure distribution, and seabed surface deformation under different load conditions, and simulation results are obtained. Subsequently, the simulation results are compared with actual data to calibrate and verify the seabed mechanics simulation model, ensuring its accuracy and reliability. If the calibration of the seabed mechanics simulation model fails, the inputs need to be adjusted until calibration and verification are passed. A simulation result dataset covering multiple working conditions is then generated and stored in the storage management module for subsequent reinforcement learning model optimization and neural network modeling and prediction. The entire process utilizes distributed computing frameworks (such as Apache Spark) and big data storage technology to achieve efficient processing and management of large amounts of simulation data, ensuring the high accuracy and reliability of the seabed mechanics simulation model.

[0042] By adjusting the input, the seabed mechanics simulation model can simulate the mechanical response under different seabed conditions (such as soft soil, hard soil, or multi-layered composite soil environments), generating simulation datasets covering various working conditions. Furthermore, the seabed mechanics simulation model is validated and compared with actual data to ensure the accuracy and reliability of the model's output. This seabed mechanics simulation model provides rich sample support for the training of subsequent reinforcement learning models and forms the theoretical basis for optimizing load distribution.

[0043] Step 3: Define the basic elements of the preset reinforcement learning model. Based on the seabed mechanics simulation model and simulation result dataset, construct a virtual training environment to simulate the operation of marine engineering equipment under different load distribution strategies, train the reinforcement learning model, and optimize the load distribution strategy through multiple interactions and iterations with the virtual training environment to generate the optimal load distribution scheme.

[0044] Specifically, the basic elements of a reinforcement learning model include a state space, an action space, and a reward function. The state space consists of the current load distribution of the marine engineering equipment, the seabed stress distribution, the contact pressure distribution, and the seabed surface deformation. The action space represents the strategy for adjusting the load distribution, and the reward function minimizes the seabed surface deformation and the energy consumption of the marine engineering equipment. Reinforcement learning models can be trained using Deep Q-learning or policy gradient algorithms.

[0045] In some embodiments, firstly, the basic elements of a pre-defined reinforcement learning model are defined. Next, based on a seabed mechanics simulation model and simulation result dataset, a virtual training environment is constructed to simulate the operation of marine engineering equipment under different load distribution strategies. Then, the reinforcement learning model is trained using a deep Q-learning algorithm or a policy gradient algorithm. Through multiple interactions and iterations with the virtual training environment, the load distribution strategy is optimized, gradually improving its performance, and ultimately obtaining the optimal load distribution scheme. The optimized load distribution strategy is applied to the load adjustment of actual marine engineering equipment, and continuously optimized through a real-time actual data feedback mechanism, forming a closed-loop iterative optimization process. The entire reinforcement learning optimization process utilizes massive simulation data stored in big data storage to ensure the comprehensiveness of model training and the efficiency of strategy optimization, fully leveraging the key role of big data technology in data-driven decision-making. Through the adaptive capability of reinforcement learning, this invention can dynamically adjust load distribution parameters to cope with complex and ever-changing seabed environments.

[0046] Step 4: Using physical parameters and load distribution of marine engineering equipment as input parameters and mechanical response data as output parameters, construct a neural network, train the neural network using the backpropagation algorithm, and predict the seabed response results under different load conditions through the trained neural network.

[0047] Specifically, neural networks include multi-layer feedforward neural networks or convolutional neural networks.

[0048] In some embodiments, a neural network is first constructed, including an input layer, multiple hidden layers, and an output layer. The input layer receives physical parameters of seabed soil samples and load distributions from marine engineering equipment; the output layer generates predictions of seabed response under different load conditions, including seabed stress distribution, contact pressure distribution, and seabed surface deformation. The neural network is then trained using an error backpropagation algorithm to optimize its parameters and introduce regularization terms to avoid overfitting, ensuring the network's generalization ability and prediction accuracy. After training, the neural network is validated using an independent validation dataset (real-world data) to ensure the prediction accuracy error is controlled within 5%. If the neural network performs poorly, it needs to return to training for parameter adjustment. After validation, a trained and validated neural network is obtained, enabling rapid prediction of seabed response under different load conditions. The entire neural network modeling and prediction process utilizes massive simulation data stored in big data storage, combined with distributed computing frameworks (such as Apache Spark and TensorFlow distributed training), to achieve efficient learning and accurate prediction of complex nonlinear relationships, fully demonstrating the crucial role of big data technology in deep learning model training and optimization.

[0049] Preferably, through training, the prediction error of the neural network is controlled within 5%, and the computational efficiency is improved by tens of times compared with traditional mechanical simulation. This neural network provides an efficient and reliable computational tool for real-time optimization, significantly shortening the response time of load optimization.

[0050] Step 5: Based on the seabed response results and the optimal load distribution scheme, dynamically adjust the load distribution of marine engineering equipment.

[0051] It should be noted that the load distribution strategy of marine engineering equipment is dynamically adjusted based on the predicted seabed response and optimal load distribution scheme to ensure that the equipment is always in optimal operating condition. Furthermore, through a closed-loop feedback mechanism, the operating strategy of the marine engineering equipment can be automatically adjusted based on real-time data to adapt to changes in the seabed soil, path, and other environmental factors, reducing damage to the seabed and improving the operational stability of the equipment. Multi-objective optimization can be achieved, minimizing seabed deformation and reducing environmental impact while maximizing energy utilization efficiency, thereby improving overall operational efficiency. Combining the advantages of neural networks and reinforcement learning models, real-time responses are possible in complex dynamic environments, ensuring efficient and stable operation of marine engineering equipment under various working conditions.

[0052] Based on the same inventive concept, another embodiment of the present invention provides an adaptive seabed load optimization device, which corresponds to the method of the foregoing embodiments, such as... Figure 2 As shown, the device includes:

[0053] The data acquisition module is used to acquire physical parameters of seabed soil samples, load distribution of marine engineering equipment, and mechanical response data during the interaction between marine engineering equipment and the seabed.

[0054] The mechanical simulation module is used to construct a seabed mechanical simulation model by taking physical parameters and load distribution of marine engineering equipment as inputs and mechanical response data as outputs. It simulates the interaction between marine engineering equipment and the seabed under different load conditions and obtains a simulation result dataset.

[0055] The reinforcement learning optimization module is used to define the basic elements of the preset reinforcement learning model. Based on the seabed mechanics simulation model and simulation result dataset, it constructs a virtual training environment to simulate the operation of marine engineering equipment under different load distribution strategies, trains the reinforcement learning model, and optimizes the load distribution strategy through multiple interactions and iterations with the virtual training environment to generate the optimal load distribution scheme.

[0056] The modeling and prediction module is used to construct a neural network with physical parameters and load distribution of marine engineering equipment as input parameters and mechanical response data as output parameters. The neural network is trained using the backpropagation algorithm and the trained neural network is used to predict the seabed response under different load conditions.

[0057] The adaptive module is used to dynamically adjust the load distribution of marine engineering equipment based on the seabed response results and the optimal load distribution scheme.

[0058] In summary, the adaptive seabed load optimization method and apparatus provided by this invention integrates big data, reinforcement learning, and neural network technologies to address the load distribution optimization problem during the interaction between marine engineering equipment and the seabed, thereby reducing seabed deformation and improving the operational stability and adaptability of marine engineering equipment. The method includes: First, acquiring seabed soil samples, classifying different types of seabed soil, and detecting their physical parameters and mechanical response data. Preprocessing the collected data using big data technology, including data cleaning, feature extraction, and standardization, to form a high-quality training and analysis dataset. Based on the collected data, a seabed mechanics simulation model is constructed using simulation software to simulate the stress distribution, deformation patterns, and interaction process between the equipment and the seabed of marine engineering equipment such as tracked vehicles and sleds. A large amount of mechanical response data is obtained through simulation analysis, providing fundamental data support for model training and optimization. On this basis, a reinforcement learning algorithm is introduced, designing state, action, and reward functions. Based on parameters such as the load distribution of marine engineering equipment and seabed deformation, the load parameters are adaptively adjusted using the reinforcement learning model and simulation feedback mechanism to continuously optimize the operating state of the marine engineering equipment. To further improve the computational efficiency and prediction accuracy of the model, neural network technology is combined with simulation datasets to construct a neural network for prediction. This network learns the nonlinear mapping relationship between load distribution and seabed response, enabling rapid prediction of seabed mechanical behavior under different load conditions and reducing the time consumption of costly simulation calculations. This neural network, combined with reinforcement learning algorithms, forms a highly efficient adaptive optimization system that dynamically adjusts the load parameters of marine engineering equipment based on real-time feedback data. This ensures optimal load distribution in complex and dynamic seabed environments, minimizing seabed deformation and energy loss.

[0059] This invention breaks through the limitations of traditional mechanical models and manual parameter tuning optimization, and has high precision, high adaptability and real-time response capabilities. It is particularly suitable for engineering scenarios with diverse seabed soil and complex and ever-changing marine environments, and provides an advanced and reliable technical means for the design and application of marine engineering equipment such as underwater robots, tracked equipment, and sleds, with broad engineering application prospects.

[0060] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. It should be understood that the invention is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An adaptive seabed load optimization method, characterized in that, include: Step 1: Obtain the physical parameters of seabed soil samples, the load distribution of marine engineering equipment, and the mechanical response data during the interaction between marine engineering equipment and the seabed; Step 2: Using the physical parameters and load distribution of the marine engineering equipment as inputs and the mechanical response data as outputs, construct a seabed mechanical simulation model using simulation software to simulate the interaction between the marine engineering equipment and the seabed under different load conditions, and obtain the simulation result dataset. Step 3: Define the basic elements of the preset reinforcement learning model. Based on the seabed mechanics simulation model and simulation result dataset, construct a virtual training environment to simulate the operation of the marine engineering equipment under different load distribution strategies, train the reinforcement learning model, and optimize the load distribution strategy through multiple interactions and iterations with the virtual training environment to generate the optimal load distribution scheme. Step 4: Using the physical parameters and the load distribution of marine engineering equipment as input parameters and the mechanical response data as output parameters, construct a neural network, train the neural network using the backpropagation algorithm, and predict the seabed response results under different load conditions through the trained neural network. Step 5: Based on the seabed response results and the optimal load distribution scheme, dynamically adjust the load distribution of the marine engineering equipment.

2. The adaptive seabed load optimization method according to claim 1, characterized in that, The physical parameters include soil density, internal friction angle, cohesion, and compression modulus, while the mechanical response data include seabed stress distribution, contact pressure distribution, and seabed surface deformation.

3. The adaptive seabed load optimization method according to claim 1, characterized in that, The marine engineering equipment includes underwater robots, tracked equipment, and sleds.

4. The adaptive seabed load optimization method according to claim 2, characterized in that, Step 1 further includes: cleaning the physical parameters and mechanical response data, removing noise and outliers, extracting key variables, and performing standardization processing, and then storing them in a data lake and data warehouse, and using a distributed storage system to realize data storage and management.

5. The adaptive seabed load optimization method according to claim 1, characterized in that, In step 2, a seabed mechanics simulation model is constructed using the discrete element method or the finite element method, and the simulation software used is EDEM or ABAQUS.

6. The adaptive seabed load optimization method according to claim 2, characterized in that, In step 3, the basic elements include state space, action space, and reward function. The state space consists of the current load distribution of the marine engineering equipment, seabed stress distribution, contact pressure distribution, and seabed surface deformation. The action space consists of the load distribution adjustment strategy. The reward function is to minimize the seabed surface deformation and the energy consumption of the marine engineering equipment.

7. The adaptive seabed load optimization method according to claim 6, characterized in that, In step 3, the reinforcement learning model is trained using a deep Q-learning algorithm or a policy gradient algorithm.

8. The adaptive seabed load optimization method according to claim 1, characterized in that, The neural network includes a multilayer feedforward neural network or a convolutional neural network.

9. The adaptive seabed load optimization method according to claim 1, characterized in that, The prediction accuracy of the neural network is controlled within 5%.

10. An adaptive seabed load optimization device, characterized in that, include: The data acquisition module is used to acquire physical parameters of seabed soil samples, load distribution of marine engineering equipment, and mechanical response data during the interaction between marine engineering equipment and the seabed. The mechanical simulation module is used to construct a seabed mechanical simulation model by taking the physical parameters and load distribution of marine engineering equipment as inputs and the mechanical response data as outputs, through simulation software, to simulate the interaction between marine engineering equipment and the seabed under different load conditions, and to obtain a simulation result dataset. The reinforcement learning optimization module is used to define the basic elements of the preset reinforcement learning model, construct a virtual training environment based on the seabed mechanics simulation model and simulation result dataset, simulate the operation of the marine engineering equipment under different load distribution strategies, train the reinforcement learning model, and optimize the load distribution strategy through multiple interactions and iterations with the virtual training environment to generate the optimal load distribution scheme. The modeling and prediction module is used to construct a neural network with the physical parameters and the load distribution of marine engineering equipment as input parameters and the mechanical response data as output parameters. The neural network is trained using the backpropagation algorithm and the trained neural network is used to predict the seabed response under different load conditions. An adaptive module is used to dynamically adjust the load distribution of marine engineering equipment based on the seabed response results and the optimal load distribution scheme.

Images