Simulation method for interactive collaborative operation of rudder propeller and thruster system

Abstract

The present invention discloses a simulation method for interactive collaborative operation of a laying ship's rudder propeller and thruster system, and relates to the field of ship operation technology. In order to solve the problem in the prior art that the rudder propeller and thruster system are independently controlled and lack an effective collaborative operation strategy, resulting in low operating efficiency, and in complex sea conditions, the ship is difficult to operate and prone to safety accidents; the present invention can more accurately control the motion trajectory of the laying ship through collaborative control of the rudder propeller and thruster system, improve laying accuracy and operating efficiency, optimize the thrust distribution strategy, reduce the energy consumption of the laying ship, and thus improve economic benefits, introduce dynamic delay factors and interference compensation strategies, and realize specialized division of labor at all levels through a hierarchical control architecture, thereby improving control efficiency and accuracy, building a case database and a strategy library, realizing rapid call and adaptive correction of historical strategies, forming a closed-loop verification process, and providing reliable theoretical support and technical guarantees for actual operations.

Description

Technical Field

[0001] The present invention relates to the technical field of ship operation, and in particular to a simulation method for interactive collaborative operation of a rudder propeller and a thruster system. Background Art

[0002] When laying submarine pipelines, laying submarine cables and other marine engineering operations, the laying vessel needs to precisely control the position and heading of the hull. Traditional laying vessels are mainly operated by rudder propellers and side thrust systems. For example, the Chinese patent with the announcement number: CN103306239B discloses a system for controlling the lateral contraction of the row body, including guide rails, supports, steel cables, row bodies, mooring ropes and steel cable hooks; transverse guide rails are loaded on both sides of the laying vessel, the guide rails are fixed to the supports by welding, the supports are fixed to the steel cable hooks, the laying vessel deck and the laying vessel slide by welding, the steel cables are connected to the supports and the laying vessel by steel cable hooks, and the row body and the guide rails are fixed to form a whole by tying ropes. The laying vessel loading transverse guide rail operation system loads transverse guide rails on both sides of the laying vessel, and then connects the row body to the guide rails through tying ropes, thereby controlling the lateral contraction of the row body and improving the effective utilization rate of the overlap width of the above-water row body. It is widely applicable to underwater laying construction under different working conditions, while reducing construction costs and improving construction efficiency.

[0003] In existing technology, the rudder propeller system is responsible for steering the ship, while the thruster system is responsible for lateral movement. In actual operations, the rudder propeller and thruster systems often need to work together to improve efficiency and precision. However, the independent control of the rudder propeller and thruster systems lacks an effective collaborative operation strategy, resulting in low efficiency. In complex sea conditions, the ship is difficult to operate and is prone to safety accidents. Summary of the Invention

[0004] The purpose of the present invention is to provide a simulation method for the interactive collaborative operation of the rudder propeller and the thruster system of the laying ship, realize the interactive collaborative operation of the rudder propeller and the thruster system, improve the control efficiency and accuracy of the laying ship, reduce the safety risks in the actual operation process through simulation, and provide a theoretical basis and technical support for the optimization of the rudder propeller and the thruster system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] The simulation method for interactive collaborative operation of rudder propeller and thruster system includes the following steps:

[0007] Step 1: Establish a dynamic model of the rudder propeller and thruster system of the laying ship, and couple the dynamic model with the control system model to generate a rudder propeller-thruster joint dynamic model. The thrust output of the rudder propeller and thruster is used as the input signal of the control system model, and the control command of the control system model is used as the input signal of the dynamic model to realize the interaction between the dynamic model and the control system model.

[0008] Step 2: Build a multi-device interactive collaborative control architecture model based on the dynamic model and control system model, and define thrust distribution constraint rules and interference compensation strategies based on the collaborative control relationship between the rudder propellers and thrusters;

[0009] Step 3: Set the simulation scenario and simulation parameters, and perform simulation calculations based on the simulation scenario and simulation parameters to simulate the motion and force conditions of the laying ship, rudder propeller, and thruster system;

[0010] Step 4: Based on the thrust distribution constraint rules and interference compensation strategy, the thrust distribution weight and control response priority are adjusted in the simulation cycle. The motion state and force conditions during the simulation are verified by deviation spectrum analysis with the preset path.

[0011] Step 5: Generate a visual simulation report based on the analysis and verification results, and mark the key collaborative control nodes and performance bottlenecks. At the same time, build a case database to store the optimized parameter combinations of typical operation scenarios, and use a similarity matching algorithm to quickly call and adaptively correct historical strategies.

[0012] Furthermore, the dynamic model in step 1 specifically includes:

[0013] Collect the design drawings of the laying ship and build a three-dimensional dynamic model of the hull, propeller and thruster system based on the principles of fluid dynamics;

[0014] Based on the three-dimensional dynamic model of the rudder propeller, the nonlinear variation characteristics of the rudder propeller thrust vector with the rudder angle and rotation speed are obtained. Based on the nonlinear variation characteristics, a rudder propeller thrust vector model is constructed. The rudder propeller thrust vector model is coupled with the hull model to simulate the influence of the rudder propeller on the hull motion.

[0015] Based on the three-dimensional dynamic model of the thrust system, the thrust data of the thrust system at different speeds are obtained, the relationship between the thrust and speed of the thrust system is established, and a discretized thrust model is formed.

[0016] Furthermore, the discretized thrust model also includes introducing a dynamic delay factor to characterize the instantaneous lag effect of propeller speed and thrust, specifically including:

[0017] Conduct at least one actual test of the thrust system, collect time series data of speed changes and thrust response under different operating conditions, and determine the dynamic delay factor τ;

[0018] The thrust model calculation formula is as follows:

[0019]

[0020] Where, T ( t ) is expressed as the thrust value at time t; T ( t-Δt ) represents the thrust value at the previous moment; ΔT ( t ) is expressed as the thrust change; T new ( n ( t )) indicates the current speed n ( t ) thrust value calculated without considering the delay effect; Δt It represents the time step of simulation calculation; τ represents the dynamic delay factor.

[0021] Furthermore, the multi-device interactive collaborative control architecture model is constructed in step 2, specifically including:

[0022] Establish a hierarchical control architecture model, including path planning layer, motion control layer and actuator layer;

[0023] The path planning layer collects environmental data in real time, generates a reference trajectory for the ship's motion based on a preset layout path, fuses the environmental data through an environmental disturbance observer, calculates the motion trajectory compensation, and compensates the reference trajectory for the ship's motion.

[0024] The motion control layer spatially synthesizes the propeller thrust vector output by the dynamic model and the thrust of the thrust system through the conversion matrix between the hull coordinate system and the local equipment coordinate system to generate a unified control input vector;

[0025] The motion control layer is also used to input the compensated hull motion reference trajectory into the rudder-propeller-thruster joint dynamics model to predict the future state of the hull and optimize the unified control input vector based on the prediction result;

[0026] The actuator layer calculates the resultant torque deviation between the rudder propeller and the thruster in real time based on the discretized thrust model, and feeds it back to the motion control layer for thrust redistribution.

[0027] Furthermore, the thrust distribution constraint rules include: a thrust direction conflict detection mechanism, a power distribution weight function, a thrust limit constraint rule and a torque balance constraint rule.

[0028] Furthermore, the interference compensation strategy includes: wake interference compensation strategy, environmental interference feedforward compensation strategy, model prediction error feedback compensation strategy and dynamic delay compensation strategy.

[0029] Furthermore, the step three also includes: configuring the ship load distribution parameters, configuring the operation task path planning data, establishing the equipment status monitoring logic, collecting the rudder angle, speed, and thrust deviation in real time as simulation feedback signals, monitoring and controlling the status of the rudder propeller and thrust system, and ensuring that the control strategy in the simulation process can be effectively executed.

[0030] Furthermore, the simulation scenario in step 3 is generated based on the actual operation type of the ship, water conditions, and equipment status, specifically including:

[0031] According to different layout training subjects, combined with corresponding layout strategies, layout characteristics are determined and a preliminary task framework is established;

[0032] Retrieve the corresponding water area map based on the paving location, and determine the environmental parameters corresponding to the paving operation training subject based on the characteristics of the water area map;

[0033] Load environmental parameters, water map, and paving locations into the preliminary task framework to form an operation scenario framework;

[0034] Obtain the key features of the equipment, build the equipment operation scenario based on the key features of the equipment, and superimpose the equipment operation scenario with the operation scenario framework to form a complete simulation scenario;

[0035] Perform three-dimensional modeling on the complete simulation scene to generate a first simulation scene, perform pre-simulation in the first simulation scene, obtain simulation data sequence, and establish a data cycle analysis mechanism;

[0036] Analyze abnormal data points that appear during the simulation process based on the data loop analysis mechanism to determine the abnormal situation and impact level corresponding to the abnormal data;

[0037] Evaluate the reliability and accuracy of the first simulation scenario based on the abnormal data and its impact, and obtain an evaluation score;

[0038] Compare the evaluation score with a preset evaluation threshold to determine whether the first simulation scenario needs to be optimized. When the evaluation score is equal to or greater than the preset evaluation threshold, it is determined that the first simulation scenario meets the requirements and does not need to be optimized.

[0039] Otherwise, the optimization criteria are determined according to the target difference between the evaluation score and the preset evaluation threshold, the optimization strategy is set, the first simulation scenario is optimized, and the second simulation scenario is generated.

[0040] Furthermore, the method further comprises:

[0041] Collect the ship's draft, water velocity, water direction and weather parameters, including wind speed and direction;

[0042] Based on the 3D dynamic model of the ship, and taking into account the draft, water velocity, water direction, and weather parameters, a simulation analysis of the interactive and coordinated operation of the rudder and propeller systems of the laying ship was carried out. Based on the simulation analysis results, a first plan for the thrust and rudder control of the laying ship was formed;

[0043] A reinforcement learning model is trained using the Deep Q-Network (DQN) algorithm, and the optimal collaborative control strategy is obtained through interactive learning with the environment. Draft, current velocity, current direction, and weather parameters are used as input data for the trained reinforcement learning model, which then outputs a collaborative control strategy. Based on this collaborative control strategy, a second solution for the thrust and propeller control of the laying vessel is obtained.

[0044] The corresponding parameters of the laying vessel thrust and rudder propeller control are extracted from the first scheme and the second scheme, and the corresponding parameters are compared. If the deviation of the corresponding parameters of the two does not exceed the set corresponding parameter threshold, the first scheme or the second scheme is adopted for control; if the deviation of the corresponding parameters of the two exceeds the set corresponding parameter threshold, the confidence evaluation of the first scheme and the second scheme is performed respectively, and the scheme with higher corresponding confidence is adopted as the control scheme.

[0045] Furthermore, the confidence assessment specifically includes:

[0046] Defining parameter confidence assessment indicators for the relevant parameters of the first solution and the second solution respectively, wherein the confidence assessment indicators include data source confidence and external link scale confidence, and the relevant parameters include source parameters and external link parameters;

[0047] The confidence values of the first or second solution are calculated using the following formulas:

[0048]

[0049] Where C is the confidence value of the first solution or the second solution, k i For the first or second option i The indicator weight of each source parameter, n is the number of source parameters of the first or second solution, C 1i For the first or second option i The data source confidence level corresponding to each source parameter; m The number of external link parameters for the first or second solution. oh jis the index weight of the j-th external link parameter of the first or second solution, C 2j The confidence level of the external link scale corresponding to the j-th external link parameter of the first or second solution.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] By collaboratively controlling the rudder propellers and side thrust systems, the motion trajectory of the laying vessel can be controlled more accurately, thereby improving the laying accuracy and operating efficiency, optimizing the thrust distribution strategy, reducing the energy consumption of the laying vessel, and thus improving economic benefits. The introduction of dynamic delay factors and interference compensation strategies can more realistically simulate actual operating conditions, and through a hierarchical control architecture, professional division of labor at each level can be achieved, improving control efficiency and accuracy. A case database and strategy library can be built to achieve rapid calling and adaptive correction of historical strategies, forming a closed-loop verification process, and providing reliable theoretical support and technical guarantees for actual operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of the simulation method for the interactive collaborative operation of the rudder propeller and thrust system of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0054] In order to solve the technical problems in existing technologies, such as the independent control of the rudder propeller and the thruster system, the lack of effective collaborative operation strategy, resulting in low operation efficiency, and the difficulty of ship operation in complex sea conditions, which is prone to safety accidents, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0055] The simulation method for interactive collaborative operation of rudder propeller and thruster system includes the following steps:

[0056] Step 1: Establish a dynamic model of the rudder propeller and thruster system of the layout ship, specifically including the rudder propeller thrust vector model and the thruster system discretized thrust model. The dynamic model is then coupled with the control system model to generate a rudder propeller-thruster joint dynamic model. The thrust output of the rudder propeller and thruster is used as the input signal of the control system model, and the control command of the control system model is used as the input signal of the dynamic model, thus realizing the interaction between the dynamic model and the control system model.

[0057] Step 2: Build a multi-device interactive collaborative control architecture model based on the dynamic model and control system model, and define thrust distribution constraint rules and interference compensation strategies based on the collaborative control relationship between the rudder propellers and thrusters;

[0058] Step 3: Set the simulation scenario and simulation parameters, including environmental parameters, initial state parameters, control parameters, etc., and perform simulation calculations based on the simulation scenario and simulation parameters to simulate the motion state and force conditions of the laying ship, rudder propeller, and thruster system;

[0059] The step three also includes: configuring ship load distribution parameters, including cabin cargo weight distribution, fuel tank liquid level data (collected through in-tank liquid level sensors), and ballast water distribution (based on real-time data from the ballast system); configuring operation task path planning data, including a layout path coordinate sequence, operation speed profile, and turning point information (imported through the task planning system); generating a dynamic reference trajectory in combination with environmental parameters to ensure that the simulation results can reflect the ship's performance under actual operating conditions; establishing equipment status monitoring logic, collecting rudder angle, speed, and thrust deviation in real time as simulation feedback signals, monitoring and controlling the status of the rudder propeller and thruster system, and ensuring that the control strategy during the simulation process can be effectively executed;

[0060] Step 4: Based on the thrust allocation constraint rules and interference compensation strategy, the thrust allocation weights and control response priorities are adjusted during the simulation cycle. The optimal solution set that satisfies the layout accuracy and energy consumption constraints is iteratively calculated. The motion state and force conditions during the simulation are compared with the preset path through deviation spectrum analysis to verify the effectiveness of the collaborative control strategy in suppressing nonlinear interference.

[0061] Step 5: Generate a visual simulation report based on the analysis and verification results, and output a multi-dimensional visual report including the spatiotemporal motion trajectory, equipment operating condition thermal map, and energy consumption distribution, and mark the key collaborative control nodes and performance bottlenecks. At the same time, build a case database to store the optimized parameter combinations of typical operating scenarios, covering dynamic model-related parameters, thrust distribution weights, control parameters, etc. Use a similarity matching algorithm to quickly call and adaptively correct historical strategies, compare the new simulation results with historical cases, further improve the strategy library, and complete the simulation closed-loop verification process.

[0062] In this embodiment, a dynamic model is established and coupled with a control system model to generate a rudder-thrust joint dynamics model, thereby achieving coordinated control of the rudder and thrust systems. Based on a multi-device interactive collaborative control architecture model, thrust distribution constraints and interference compensation strategies are defined to address the low operating efficiency caused by independent control of the rudder and thrust systems. By configuring ship load distribution parameters, task path planning data, and environmental parameters, and combining equipment status monitoring logic to collect feedback signals in real time, the simulation results are ensured to accurately reflect the ship's performance under actual operating conditions. In the simulation cycle, based on the thrust distribution constraints and interference compensation strategies, the thrust distribution weights and control response priorities are dynamically adjusted, and the optimal solution set that meets the layout accuracy and energy consumption constraints is iteratively calculated. The deviation spectrum analysis verifies the collaborative control strategy's suppression effect on nonlinear interference. Finally, a multi-dimensional visual simulation report is generated, a case database is constructed, and a similarity matching algorithm is used to achieve rapid recall and adaptive correction of historical strategies, forming a closed-loop verification process. This significantly improves the maneuvering accuracy and safety of the layout vessel in complex sea conditions, reduces energy consumption, and provides reliable theoretical support and technical guarantees for actual operations.

[0063] In this embodiment, the dynamic model in step 1 specifically includes:

[0064] The design drawings of the layout ship were collected, including information such as the ship's line drawings, cabin layout, and the installation locations and dimensions of the rudder propellers and thrusters. This information was used to determine the model's outline and the relative positions of its components. Based on the principles of ship engineering, the hull was abstracted as a rigid body, and its physical parameters such as mass, center of gravity, and moment of inertia were defined. A three-dimensional dynamic model of the hull, rudder propeller, and thruster system was constructed based on the principles of fluid dynamics. This model simulated the movement and forces of the hull in water, described the geometric characteristics of the propeller blades, such as shape, pitch, and diameter, and characterized the nonlinear variation of the rudder propeller thrust vector with rudder angle and rotational speed.

[0065] Based on the three-dimensional dynamic model of the rudder propeller, the nonlinear variation characteristics of the rudder propeller thrust vector with the rudder angle and rotation speed are obtained. Based on the nonlinear variation characteristics, a rudder propeller thrust vector model is constructed. The rudder propeller thrust vector model is coupled with the hull model to simulate the influence of the rudder propeller on the hull motion.

[0066] Based on the three-dimensional dynamic model of the thrust system, the thrust data of the thrust system at different speeds is obtained, and the relationship between the thrust and speed of the thrust system is established to form a discretized thrust model. The discretized thrust model also includes the introduction of a dynamic delay factor to characterize the instantaneous lag effect between the propeller speed and thrust, specifically including:

[0067] Conduct at least one actual test of the thrust system, collect time series data on speed changes and thrust response under different operating conditions, and determine the dynamic delay factor τ related to factors such as propeller characteristics, fluid viscosity, and system response time;

[0068] The thrust model calculation formula is as follows:

[0069]

[0070] Where, T ( t ) is expressed as the thrust value at time t; T ( t-Δt ) represents the thrust value at the previous moment; ΔT ( t ) is expressed as the thrust change; T new ( n ( t )) indicates the current speed n ( t ) thrust value calculated without considering the delay effect; Δt It represents the time step of simulation calculation; τ represents the dynamic delay factor.

[0071] In this embodiment, by collecting drawing information to construct a three-dimensional dynamic model, the movement and force of the hull and equipment in the water can be accurately simulated, providing a reliable basis for subsequent simulations. The rudder propeller thrust vector model and the discretized thrust model effectively characterize the equipment characteristics. By introducing a dynamic delay factor, the instantaneous lag effect of the propeller speed and thrust is more realistically reflected, making the simulation results more in line with actual operating conditions, greatly improving the accuracy and practicality of the simulation.

[0072] In this embodiment, the multi-device interactive collaborative control architecture model is constructed in step 2, specifically including:

[0073] Establish a hierarchical control architecture model, including path planning layer, motion control layer and actuator layer;

[0074] The path planning layer collects environmental data such as wind speed, wave direction angle, and current velocity in real time through onboard wind sensors, wave height meters, and current meters. It generates a reference trajectory for the ship's motion based on the preset layout path, integrates the environmental data through an environmental disturbance observer, calculates the motion trajectory compensation, and compensates the reference trajectory for the ship's motion.

[0075] The motion control layer spatially synthesizes the propeller thrust vector output by the dynamic model and the thrust of the thrust system through the conversion matrix between the hull coordinate system and the local equipment coordinate system to generate a unified control input vector;

[0076] The motion control layer is also used to input the compensated hull motion reference trajectory into the rudder-propeller-thruster joint dynamic model to predict the future state of the hull and optimize the unified control input vector based on the prediction results to achieve optimization of layout accuracy and energy consumption;

[0077] The actuator layer calculates the resultant torque deviation of the rudder propeller and the thruster in real time based on the discretized thrust model, and feeds it back to the motion control layer for thrust redistribution, so that the simulated thrust curve is consistent with the actual equipment response characteristics.

[0078] In this embodiment, the thrust distribution constraint rules include:

[0079] Thrust direction conflict detection mechanism: Based on the real-time thrust vectors of the rudder propeller and thruster, the transverse projection angle θ is calculated through the vector dot product. When |cosθ| < δ, it is determined to be a direction conflict, triggering the priority arbitration logic. The conflict threshold δ is calibrated using actual ship test data and has a value range of 0.1≤δ≤0.3;

[0080] Power allocation weight function: define dynamic weight coefficients , where n is the cumulative working time of the equipment, which is read in real time by the actuator layer PLC, n max is the maximum continuous working time of the equipment (set according to the equipment technical manual), k is the weight decay rate coefficient, determined by regression analysis of historical fault data, and automatically switches to the backup equipment when γ(n)<0.5;

[0081] Thrust limit constraint rule: Set the rudder propeller thrust limit |F0|≤F 0max (F0 max Calculated based on the rated power of the rudder propeller motor), thrust limiter |F1|≤F 1max (F 1max Determined based on hydraulic system pressure threshold);

[0082] Torque balance constraint rule: The combined torque of the rudder propeller and the thruster must satisfy ||F0×r0+F1×r1||≤τ max , where r0 and r1 are the force arm vectors extracted from the design drawing in step 1, τ max is the maximum allowable rotational moment of the hull structure.

[0083] In this embodiment, the interference compensation strategy includes:

[0084] Wake disturbance compensation strategy: The bow Doppler current meter is used to collect the disturbance of the propeller wake on the thruster inlet velocity in real time. The thrust efficiency loss is calculated based on the Bernoulli equation, and the compensation value is generated and injected into the control command.

[0085] Environmental disturbance feedforward compensation strategy: Real-time environmental data is acquired through onboard wind sensors, wave height meters, and ADCPs. The environmental disturbance force is calculated based on fluid mechanics formulas and added to the control input vector as a feedforward term.

[0086] Model prediction error feedback compensation strategy: The actual resultant torque deviation Δτ is calculated in real time through the thrust space coupling observer at the actuator layer. When Δτ is greater than the allowable error threshold, the adaptive compensation algorithm is triggered to correct the thrust distribution weights α and β.

[0087] Dynamic delay compensation strategy: A dynamic delay factor τ is embedded in the discretized thrust model to correct the thrust output timing in real time.

[0088] In this embodiment, the hierarchical control architecture model realizes the specialized division of labor at each level. The path planning layer accurately plans the trajectory based on environmental data, the motion control layer optimizes the control input, and the actuator layer ensures that the thrust curve is consistent with the actual situation, thereby improving the efficiency and accuracy of the control. The thrust distribution constraint rules can effectively avoid thrust conflicts, reasonably distribute power, ensure the safe operation of the equipment, and maintain the hull torque balance; the interference compensation strategy compensates and corrects interference factors such as wake, environment, model error and dynamic delay in real time, thereby improving the stability and reliability of the system under complex working conditions and greatly enhancing the practicality and accuracy of the layout operation simulation.

[0089] In this embodiment, the simulation scenario in step 3 is generated based on the actual operation type of the ship, water conditions, and equipment status, and specifically includes:

[0090] According to different deployment training subjects, such as routine deployment, complex terrain deployment, and emergency rescue deployment, deployment characteristics are determined in combination with corresponding deployment strategies to establish a preliminary operational task framework. For example, routine deployment focuses on the balance between deployment speed and accuracy; complex terrain deployment requires key consideration of the terrain's impact on ship posture and equipment operation; and emergency rescue deployment emphasizes rapid response and efficient coordination.

[0091] Retrieve the corresponding water map based on the paving location, such as maps of ports, inland rivers, offshore areas, and other different water areas. Based on the characteristics of the water map, such as channel width, water flow characteristics, and seabed topography, determine the environmental parameters corresponding to the paving operation training subjects, including water flow speed, direction, wave height, tidal changes, etc.

[0092] Load environmental parameters, water map, and paving locations into the preliminary task framework to form an operation scenario framework;

[0093] Obtain key characteristics of the equipment, such as the thrust characteristics of the rudder propeller and the response time of the thrust system. Build an equipment operation scenario based on these key characteristics and overlay the equipment operation scenario with the operation scenario framework to form a complete simulation scenario.

[0094] Perform three-dimensional modeling on the complete simulation scene to generate a first simulation scene, perform pre-simulation in the first simulation scene, obtain simulation data sequence, and establish a data cycle analysis mechanism;

[0095] Analyze abnormal data points that appear during the simulation process based on the data loop analysis mechanism to determine the abnormal situation and impact level corresponding to the abnormal data, such as equipment response delay, force calculation deviation, etc.

[0096] Evaluate the reliability and accuracy of the first simulation scenario based on the abnormal data and its impact, and obtain an evaluation score;

[0097] Compare the evaluation score with a preset evaluation threshold to determine whether the first simulation scenario needs to be optimized. When the evaluation score is equal to or greater than the preset evaluation threshold, it is determined that the first simulation scenario meets the requirements and does not need to be optimized.

[0098] Otherwise, the optimization criteria are determined according to the target difference between the evaluation score and the preset evaluation threshold, the optimization strategy is set, the first simulation scenario is optimized, and the second simulation scenario is generated.

[0099] In this embodiment, the diversity of laying operations, the complexity of the water environment and the characteristics of the equipment are fully considered according to the actual operation type, water conditions and equipment status, so that the constructed simulation scene is highly close to the actual operation situation, providing a reliable realistic basis for subsequent simulation calculations. Through pre-simulation, data loop analysis and comparison with preset evaluation thresholds, abnormal situations in the scene can be discovered in time and optimized in a targeted manner, thereby improving the reliability and accuracy of the simulation scene, thereby ensuring that the simulation results based on the scene are more valuable for reference, and providing strong support for the research and training of laying ship operations.

[0100] Based on the above embodiment, the method further includes:

[0101] Collect the ship's draft, water velocity, water direction and weather parameters, including wind speed and direction;

[0102] Based on the 3D dynamic model of the ship, and taking into account the draft, water velocity, water direction, and weather parameters, a simulation analysis of the interactive and coordinated operation of the rudder and propeller systems of the laying ship was carried out. Based on the simulation analysis results, a first plan for the thrust and rudder control of the laying ship was formed;

[0103] A reinforcement learning model is trained using the Deep Q-Network (DQN) algorithm, and the optimal collaborative control strategy is obtained through interactive learning with the environment. Draft, current velocity, current direction, and weather parameters are used as input data for the trained reinforcement learning model, which then outputs a collaborative control strategy. Based on this collaborative control strategy, a second solution for the thrust and propeller control of the laying vessel is obtained.

[0104] The corresponding parameters of the laying vessel thrust and rudder propeller control are extracted from the first scheme and the second scheme, and the corresponding parameters are compared. If the deviation of the corresponding parameters of the two does not exceed the set corresponding parameter threshold, the first scheme or the second scheme is adopted for control; if the deviation of the corresponding parameters of the two exceeds the set corresponding parameter threshold, the confidence evaluation of the first scheme and the second scheme is performed respectively, and the scheme with higher corresponding confidence is adopted as the control scheme.

[0105] Specifically, by collecting the hull's draft, water velocity, water direction and weather parameters, based on the hull's three-dimensional dynamic model, interactive collaborative operation simulation analysis is carried out to form the first scheme for the thrust and rudder propeller control of the laying ship; in addition, the deep Q-network (DQN) algorithm is used to train the reinforcement learning model, and the reinforcement learning model outputs the collaborative control strategy to obtain the second scheme for the thrust and rudder propeller control of the laying ship; the corresponding parameters related to the thrust and rudder propeller control of the laying ship in the two schemes are determined and compared. If the deviations are all within an acceptable range, it means that the analysis results of different methods are basically consistent, and any execution scheme can be selected; if there are large deviations, it means that the analysis results of different methods are inconsistent, that is, there will be analysis results with insufficient accuracy. At this time, a confidence assessment is implemented to identify which scheme is credible and reliable, and the reliable scheme will be selected for execution; this scheme examines the reliability of two different schemes by implementing a confidence assessment, thereby ensuring the execution of a reliable scheme. The adoption of this scheme is conducive to improving control accuracy.

[0106] Based on the above embodiment, the confidence evaluation specifically includes:

[0107] Defining parameter confidence assessment indicators for relevant parameters of the first solution and the second solution respectively, wherein the confidence assessment indicators include data source confidence and external link scale confidence, and the relevant parameters include source parameters and external link parameters;

[0108] The confidence values of the first or second solution are calculated using the following formulas:

[0109]

[0110] Where C is the confidence value of the first solution or the second solution, k i For the first or second optioni The indicator weight of each source parameter, n is the number of source parameters of the first or second solution, C 1i For the first or second option i The data source confidence level corresponding to each source parameter; m The number of external link parameters for the first or second solution. oh j is the index weight of the j-th external link parameter of the first or second solution, C 2j The confidence level of the external link scale corresponding to the j-th external link parameter of the first or second solution.

[0111] Specifically, this solution provides a usable solution confidence assessment method. On the basis of considering the confidence of the data source and the confidence of the external link scale, the above formula is used to calculate the confidence assessment value to obtain the comprehensive confidence value of two different solutions, thereby determining the reliability of the solution; this solution is based on the confidence assessment of different parameters in the same solution, and is combined with the parameter weight data in the assessment to calculate a more reasonable comprehensive confidence value; wherein, the sum of the indicator weights of all source parameters and the indicator weights of the external link parameters used in the calculation is equal to 1; the present invention adopts this solution to achieve high-precision and reliable collaborative operation simulation analysis.

[0112] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. The simulation method for interactive collaborative operation of rudder propeller and thruster system is characterized by: The following steps are involved: A dynamic model of the rudder propeller and thruster system of the laying ship is established, and the dynamic model is coupled with the control system model to generate a rudder propeller-thruster joint dynamic model. The thrust output of the rudder propeller and thruster is used as the input signal of the control system model, and the control command of the control system model is used as the input signal of the dynamic model. A multi-device interactive collaborative control architecture model is constructed based on the dynamic model and control system model. The thrust distribution constraint rules and interference compensation strategy are defined according to the collaborative control relationship between the rudder propeller and the thruster. Set simulation scenarios and simulation parameters, perform simulation calculations based on the simulation scenarios and simulation parameters, and simulate the motion state and force conditions of the laying ship, rudder propeller and thrust system; Based on the thrust distribution constraint rules and interference compensation strategy, the thrust distribution weight and control response priority are adjusted in the simulation cycle, and the motion state and force conditions during the simulation are verified by deviation spectrum analysis with the preset path; Based on the analysis and verification results, a visual simulation report is generated, and key collaborative control nodes and performance bottlenecks are marked. A case database is built to store optimized parameter combinations for typical operation scenarios, and a similarity matching algorithm is used to quickly call and adaptively correct historical strategies.

2. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 1, characterized in that: Dynamic models, specifically including: Collect the design drawings of the laying ship and build a three-dimensional dynamic model of the hull, propeller and thruster system based on the principles of fluid dynamics; Based on the three-dimensional dynamic model of the rudder propeller, the nonlinear variation characteristics of the rudder propeller thrust vector with the rudder angle and rotation speed are obtained. Based on the nonlinear variation characteristics, a rudder propeller thrust vector model is constructed. The rudder propeller thrust vector model is coupled with the hull model to simulate the influence of the rudder propeller on the hull motion. Based on the three-dimensional dynamic model of the thrust system, the thrust data of the thrust system at different speeds are obtained, the relationship between the thrust and speed of the thrust system is established, and a discretized thrust model is formed.

3. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 2, characterized in that: The discretized thrust model also includes introducing a dynamic delay factor to characterize the instantaneous lag effect of propeller speed and thrust, specifically including: Conduct at least one actual test of the thrust system, collect time series data of speed changes and thrust response under different operating conditions, and determine the dynamic delay factor τ; The thrust model calculation formula is as follows: Where, T ( t ) is expressed as the thrust value at time t; T ( t-Δt ) represents the thrust value at the previous moment; ΔT ( t ) is expressed as the thrust change; T new ( n ( t )) indicates the current speed n ( t ) thrust value calculated without considering the delay effect; Δt It represents the time step of simulation calculation; τ represents the dynamic delay factor.

4. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 3, characterized in that: Construct a multi-device interactive collaborative control architecture model, specifically including: Establish a hierarchical control architecture model, including path planning layer, motion control layer and actuator layer; The path planning layer collects environmental data in real time, generates a reference trajectory for the ship's motion based on a preset layout path, fuses the environmental data through an environmental disturbance observer, calculates the motion trajectory compensation, and compensates the reference trajectory for the ship's motion. The motion control layer spatially synthesizes the propeller thrust vector output by the dynamic model and the thrust of the thrust system through the conversion matrix between the hull coordinate system and the local equipment coordinate system to generate a unified control input vector; The motion control layer is also used to input the compensated hull motion reference trajectory into the rudder-propeller-thruster joint dynamics model to predict the future state of the hull and optimize the unified control input vector based on the prediction result; The actuator layer calculates the resultant torque deviation between the rudder propeller and the thruster in real time based on the discretized thrust model, and feeds it back to the motion control layer for thrust redistribution.

5. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 4, characterized in that: The thrust distribution constraint rules include: a thrust direction conflict detection mechanism, a power distribution weight function, a thrust limit constraint rule and a torque balance constraint rule.

6. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 5, characterized in that: The interference compensation strategies include: wake interference compensation strategy, environmental interference feedforward compensation strategy, model prediction error feedback compensation strategy and dynamic delay compensation strategy.

7. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 6, characterized in that: Set the simulation scenario and simulation parameters, perform simulation calculations based on the simulation scenario and simulation parameters, simulate the motion state and force conditions of the laying ship, rudder propeller and thrust system, and also include: configuring the ship load distribution parameters. At the same time, configure the operation task path planning data, establish equipment status monitoring logic, collect rudder angle, speed, and thrust deviation in real time as simulation feedback signals, monitor and control the status of the rudder propeller and thrust system, and ensure that the control strategy in the simulation process can be effectively executed.

8. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 7, characterized in that: The simulation scenario is generated based on the actual ship operation type, water conditions, and equipment status, specifically including: According to different layout training subjects, combined with corresponding layout strategies, layout characteristics are determined and a preliminary task framework is established; Retrieve the corresponding water area map based on the paving location, and determine the environmental parameters corresponding to the paving operation training subject based on the characteristics of the water area map; Load environmental parameters, water map, and paving locations into the preliminary task framework to form an operation scenario framework; Obtain the key features of the equipment, build the equipment operation scenario based on the key features of the equipment, and superimpose the equipment operation scenario with the operation scenario framework to form a complete simulation scenario; Perform three-dimensional modeling on the complete simulation scene to generate a first simulation scene, perform pre-simulation in the first simulation scene, obtain simulation data sequence, and establish a data cycle analysis mechanism; Analyze abnormal data points that appear during the simulation process based on the data loop analysis mechanism to determine the abnormal situation and impact level corresponding to the abnormal data; Evaluate the reliability and accuracy of the first simulation scenario based on the abnormal data and its impact, and obtain an evaluation score; Compare the evaluation score with a preset evaluation threshold to determine whether the first simulation scenario needs to be optimized. When the evaluation score is equal to or greater than the preset evaluation threshold, it is determined that the first simulation scenario meets the requirements and does not need to be optimized. Otherwise, the optimization criteria are determined according to the target difference between the evaluation score and the preset evaluation threshold, the optimization strategy is set, the first simulation scenario is optimized, and the second simulation scenario is generated.

9. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system of the laying ship according to claim 2, characterized in that: Also includes: Collect the ship's draft, water velocity, water direction and weather parameters, including wind speed and direction; Based on the 3D dynamic model of the ship, and taking into account the draft, water velocity, water direction, and weather parameters, a simulation analysis of the interactive and coordinated operation of the rudder and propeller systems of the laying ship was carried out. Based on the simulation analysis results, a first plan for the thrust and rudder control of the laying ship was formed; A deep Q-network algorithm is used to train a reinforcement learning model, and the optimal collaborative control strategy is obtained through interactive learning with the environment. Draft, current velocity, current direction, and weather parameters are used as input data for the trained reinforcement learning model, which then outputs a collaborative control strategy. Based on this collaborative control strategy, a second solution for the thrust and propeller control of the laying vessel is obtained. The corresponding parameters of the laying vessel thrust and rudder propeller control are extracted from the first scheme and the second scheme, and the corresponding parameters are compared. If the deviation of the corresponding parameters of the two does not exceed the set corresponding parameter threshold, the first scheme or the second scheme is adopted for control; if the deviation of the corresponding parameters of the two exceeds the set corresponding parameter threshold, the confidence evaluation of the first scheme and the second scheme is performed respectively, and the scheme with higher corresponding confidence is adopted as the control scheme.

10. The method for simulating the interactive collaborative operation of the rudder propeller and thruster system according to claim 9, characterized in that: The confidence assessment specifically includes: Defining parameter confidence assessment indicators for the relevant parameters of the first solution and the second solution respectively, wherein the confidence assessment indicators include data source confidence and external link scale confidence, and the relevant parameters include source parameters and external link parameters; The confidence values of the first or second solution are calculated using the following formulas: Where C is the confidence value of the first solution or the second solution, k i For the first or second option i The indicator weight of each source parameter, n is the number of source parameters of the first or second solution, C 1i For the first or second option i The data source confidence level corresponding to each source parameter; m The number of external link parameters for the first or second solution. ω j is the index weight of the j-th external link parameter of the first or second solution, C 2j The confidence level of the external link scale corresponding to the j-th external link parameter of the first or second solution.

Images