Ultra-large container ship energy efficiency optimization system based on intelligent navigation control

Abstract

The invention relates to the technical field of container ship energy efficiency optimization, and discloses an ultra-large container ship energy efficiency optimization system based on intelligent navigation control, and the system comprises an environment sensing module which collects the multi-source data of the navigation environment and the self state in real time; the state prediction module is used for fusing data to construct a wave field and predicting a future waveform and ship response; the intelligent decision-making module is used for solving multi-objective optimization by using MPC based on the prediction information and generating a cooperative control instruction; the cooperative execution module analyzes the optimal instruction and converts the optimal instruction into a cooperative control signal of the propulsion and control system; and the efficiency feedback module is used for evaluating actual energy efficiency, constructing an efficiency map and feeding back and correcting the macroscopic route cost. According to the invention, a Kalman filtering framework in the state prediction module is utilized, and statistical prior information provided by a wide-area wave spectrum and observation input provided by a near-field high-precision sea surface elevation measurement value are deeply fused, so that high-precision and deterministic prediction of a local waveform sequence to be encountered by a ship is realized.

Description

Technical Field

[0001] This invention relates to the field of container ship energy efficiency optimization technology, specifically to an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control. Background Technology

[0002] For ultra-large container ships navigating the ocean, the core operational indicators are energy efficiency and navigation safety. Both of these indicators are directly affected by the complex and dynamic marine environment, especially sea waves.

[0003] Existing technologies primarily operate at two levels: at the macro-strategic level, ships generally employ weather navigation systems. These systems, based on wide-ranging weather and sea state forecasts and combined with a static performance baseline model of the ship, plan a theoretically optimal route for the entire voyage. At the micro-execution level, ships rely on conventional automated equipment. For example, the autopilot maintains the preset course, while the main engine speed control system maintains a constant speed or power output. These two levels of technological application are relatively independent, each responsible for different aspects of navigation.

[0004] However, existing ship energy efficiency optimization technologies, such as automatic steering and main engine speed control systems, can only compensate for changes in hull attitude or speed that have already occurred, and their control effect lags behind the actual impact of waves. Single-objective control strategies based on constant speed or constant power often lead to risks such as ship stall, sudden increases in structural loads, or propeller idling in severe sea conditions, forcing operators to drastically reduce speed to ensure safety, which severely sacrifices energy efficiency. Furthermore, the static performance models relied upon by weather and navigation systems are insufficient to determine the true energy-saving potential of advanced microcontroller systems on ships under specific sea conditions. Therefore, this invention provides an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control. This system solves the problems of existing ship energy efficiency optimization technologies, which typically rely solely on weather forecasts for route planning. These technologies are ill-suited to handle instantaneous and localized dangerous sea conditions encountered during navigation and lack real-time decision-making capabilities at the micro level. Furthermore, conventional autopilot or main engine control systems are mostly passive responses, making it difficult to proactively utilize wave information for refined energy efficiency optimization while ensuring structural and cargo safety.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control, comprising:

[0007] The environmental perception module is used to collect multi-source data on the ship's navigation environment and its own status in real time;

[0008] The state prediction module is used to receive and fuse the multi-source data to construct a local deterministic wavefield model, and to predict the waveform sequence in the short time domain and the multi-dimensional ship response caused by the waveform sequence based on the local deterministic wavefield model, so as to obtain multi-dimensional state prediction information.

[0009] The intelligent decision-making module is used to receive the multi-dimensional state prediction information, use model predictive control methods to solve the optimization problem of minimizing the multi-objective dynamic cost function, and generate the optimal cooperative control command sequence for the host and servo motor.

[0010] The collaborative execution module is used to parse the optimal collaborative control command sequence and convert it into specific control signals for the ship's propulsion system and maneuvering system to coordinate actions;

[0011] The performance feedback module is used to evaluate the actual operating performance of the collaborative execution module under different sea conditions, construct and update a dynamic performance map describing the energy-saving potential, and feed the dynamic performance map back to the ship's macro-meteorological navigation system to correct the cost function of global route optimization.

[0012] Preferably, the environment sensing module includes:

[0013] The local sea state data acquisition unit integrates an X-band navigation radar for inverting wide-area two-dimensional wave spectra and a lidar system for directly acquiring near-field instantaneous three-dimensional waveform profiles.

[0014] The ship status data acquisition unit consists of an inertial measurement unit, a global positioning system, and a propulsion system sensor group used to acquire the real-time speed, output torque, and instantaneous fuel consumption rate of the main engine;

[0015] The external environment data acquisition unit includes an anemometer for acquiring relative wind speed and relative wind direction angle;

[0016] The data synchronization and preprocessing unit is used to assign unified timestamps and perform filtering processing on the multi-source data collected by the local sea state data acquisition unit, the ship status data acquisition unit and the external environment data acquisition unit to form standardized structured data frames.

[0017] Preferably, the state prediction module includes:

[0018] The wave field reconstruction and forward prediction unit is used to fuse the multi-source data using a Kalman filter framework. Specifically, it uses the wide-area wave spectrum provided by the X-band navigation radar as statistical prior information and the near-field high-precision sea surface elevation measurement provided by the lidar system as observation input to estimate and update the state vector of the local wave field and perform forward calculation on the waveform sequence in the future prediction time domain.

[0019] The multi-dimensional ship response prediction unit is used to receive waveform sequences and, in conjunction with a pre-set ship hydrodynamic performance database, predict multi-dimensional response states in real time.

[0020] Preferably, the real-time prediction of multi-dimensional response states specifically includes:

[0021] Instantaneous fuel consumption rate related to energy efficiency;

[0022] Total longitudinal bending moment related to structural loads;

[0023] Overall acceleration of cargo position.

[0024] Preferably, the intelligent decision-making module is specifically used for:

[0025] Based on the multi-dimensional state prediction information, the multi-objective dynamic cost function is constructed. In the prediction time domain, the multi-objective dynamic cost function performs a weighted summation of multiple cost items representing operational energy efficiency, structural safety and cargo safety.

[0026] The energy efficiency cost item is determined based on the predicted instantaneous fuel consumption rate, the structural safety cost item is determined based on the predicted total longitudinal bending moment, and the cargo safety cost item is determined based on the predicted comprehensive acceleration of cargo position.

[0027] Preferably, the intelligent decision-making module includes a risk assessment and weight adaptation unit specifically used for:

[0028] The comprehensive risk level index is calculated based on the peak value of the combined acceleration of the predicted structural load and cargo position throughout the entire prediction time domain.

[0029] The dynamic weighting factor is dynamically adjusted based on the comprehensive risk level index, and the dynamic adjustment steps include:

[0030] When the comprehensive risk level index is low, the weighting factor of energy efficiency cost is increased, while the weighting factors of structural safety cost and cargo safety cost are decreased, in order to operate in an energy efficiency priority mode.

[0031] When the comprehensive risk level index increases, the weighting factors of structural safety cost and cargo safety cost are increased to automatically switch to a safety-first mode.

[0032] Preferably, the intelligent decision-making module further includes an optimal control sequence solving unit specifically used for:

[0033] A numerical optimization algorithm is used to solve the multi-objective dynamic cost function after the weights are adjusted by the risk assessment and weight adaptation unit.

[0034] The solution process follows physical constraints, which include: the maximum and minimum rotational speeds of the main engine, the power increase / decrease rate limit of the main engine, the maximum usable rudder angle of the rudder, and the maximum rudder turning rate limit.

[0035] Preferably, the collaborative actions performed by the collaborative execution module specifically include:

[0036] When the intelligent decision-making module receives a positive power adjustment command based on the prediction that the bow will encounter a large wave that will generate significant additional drag, the main engine output power is increased in advance to compensate for the wave-induced drag.

[0037] When the intelligent decision-making module receives a power reduction command based on a prediction that the ship is about to pass through a wave crest, which may increase the risk of propeller idling or cause a sudden decrease in load, it will execute an instantaneous and small-amplitude power reduction to proactively adapt to load changes.

[0038] When the intelligent decision-making module receives a rudder angle adjustment command generated to reduce resistance at a specific wave encounter angle, the rudder angle is deflected to actively generate a roll suppression torque to counteract the roll caused by the waves, or to adjust the relative attitude of the hull and the waves to reduce wave-induced additional resistance. The rudder angle deflection will not cause the ship to deviate significantly from the predetermined course.

[0039] Preferably, the performance feedback module includes:

[0040] The dynamic performance evaluation unit is used to receive the instantaneous fuel consumption rate of the ship's navigation environment and propulsion system sensors from the environmental perception module, and call the preset ship baseline performance model to obtain the benchmark fuel consumption rate under the same ship navigation environment, and calculate the energy efficiency gain factor by comparison.

[0041] The dynamic performance map construction unit is used to collect multiple data points generated by the dynamic performance evaluation unit under different ship navigation environments in order to construct a multi-dimensional dynamic performance map that is updated online and describes the energy-saving potential.

[0042] The macro-route cost function correction unit is used to provide the multi-dimensional dynamic performance map as a correction parameter to the macro-meteorological navigation system in order to correct the cost function of global route optimization.

[0043] Preferably, the pre-set ship baseline performance model is a performance reference model that describes the benchmark fuel consumption rate of a ship under a specific combination of speed, draft, wind conditions, and sea conditions, established in advance through ship tank tests, numerical simulation calculations, or actual ship navigation test data.

[0044] This invention provides an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control. It offers the following advantages:

[0045] 1. This invention integrates the X-band navigation radar and lidar system in the environmental perception module, and utilizes the Kalman filter framework in the state prediction module to deeply fuse the statistical prior information provided by the wide-area wave spectrum with the observation input provided by the near-field high-precision sea surface elevation measurement. This enables high-precision, deterministic prediction of the local waveform sequence that the ship is about to encounter, providing an input basis for subsequent predictive collaborative control.

[0046] 2. This invention utilizes an intelligent decision-making module to construct a multi-objective dynamic cost function that includes energy efficiency costs, structural safety costs, and cargo safety costs. Through risk assessment and weight adaptive units that evaluate the comprehensive risk level in real time, the system can dynamically adjust the weighting factors of each cost, automatically switching between an "energy efficiency priority mode" and a "safety priority mode." This allows for maximizing energy-saving potential while ensuring the safety of the ship's structure and cargo.

[0047] 3. By setting up an efficiency feedback module, this invention realizes an adaptive closed loop of micro-level collaborative control and macro-level route planning. By comparing the actual fuel consumption rate with the baseline model, the energy efficiency gain factor of this system under specific sea conditions is quantitatively calculated, and a dynamic efficiency map is constructed to correct the cost function of the macro-level meteorological navigation system. This changes the decision basis for global route planning from "minimum wind and waves" to "maximum energy-saving potential of this system", thereby achieving optimal energy efficiency for the entire voyage. Attached Figure Description

[0048] Figure 1 This is a system architecture diagram of the present invention;

[0049] Figure 2 This is a schematic diagram of the ship sensor of the present invention;

[0050] Figure 3 This is a schematic diagram of the intelligent decision-making module of the present invention;

[0051] Figure 4 This is a schematic diagram of the collaborative execution module of the present invention. Detailed Implementation

[0052] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] See attached document Figure 1 , Figure 1This is a system architecture diagram of an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control, according to an embodiment of the present invention. The present invention provides an energy efficiency optimization system for ultra-large container ships based on intelligent navigation control, including an environmental perception module, a state prediction module, an intelligent decision-making module, a collaborative execution module, and an efficiency feedback module.

[0054] The environmental perception module is used to collect multi-source data on the ship's navigation environment and its own status in real time. This module integrates an X-band navigation radar, a lidar system, an inertial measurement unit (IMU), a global positioning system (GPS), and propulsion system sensors. The X-band navigation radar acquires wide-area two-dimensional wave spectrum information around the ship, while the lidar system directly measures high-precision instantaneous sea surface waveform profiles in the near-field area forward of the bow. The IMU and GPS provide the ship's six-degree-of-freedom motion attitude and navigation information, while the propulsion system sensors monitor operating parameters such as the main engine's speed and torque.

[0055] The state prediction module receives and fuses multi-source sensing data, and constructs a locally deterministic wavefield model using algorithms such as Kalman filtering. Based on this locally deterministic wavefield model, the state prediction module performs forward prediction of waveform sequences in the short time domain and further predicts the multi-dimensional ship responses caused by these waveform sequences. These responses include wave-induced additional drag, propulsion system efficiency fluctuations, total longitudinal bending moment of the hull structure, and cargo acceleration at key locations.

[0056] The intelligent decision-making module receives multi-dimensional state prediction information from the state prediction module. This module constructs a multi-objective dynamic cost function encompassing energy efficiency cost, structural safety cost, and cargo safety cost. It assesses the navigation risk level based on predicted structural loads and cargo acceleration, dynamically adjusting the weighting factors of each cost item and adaptively switching between energy efficiency-first and safety-first modes. Employing model predictive control, this module solves the optimization problem of this cost function while satisfying the ship's physical constraints, generating the optimal coordinated control command sequence for the main engine and steering gear.

[0057] The collaborative execution module parses the optimal collaborative control command sequence and converts it into specific control signals for the ship's underlying equipment. This module sends power or speed adjustment commands to the main engine governor and simultaneously sends rudder angle fine-tuning commands to the autopilot system. The actuators of the main engine and steering gear coordinate their actions according to the commands, adjusting the ship's propulsion status and navigation attitude at high frequency.

[0058] The performance feedback module continuously monitors and records the actual operational energy efficiency of the collaborative execution module under different sea states, compares it with the ship's baseline performance model, and calculates energy efficiency gain factors. These factors are used to construct and update a dynamic performance map describing the system's energy-saving potential under different sea states. This map, as a correction parameter, is fed back to the ship's top-level macro-meteorological navigation system to optimize the cost function of the global route, thereby enabling future route planning to incorporate the system's actual control performance into decision-making.

[0059] See attached document Figure 2 The environmental perception module is used to acquire comprehensive and real-time internal and external environmental data required for ship navigation, and to provide a standardized data stream that has been synchronized and preprocessed for the subsequent state prediction module.

[0060] The environmental perception module includes a local sea state data acquisition unit, which integrates an X-band navigation radar and a lidar system. The X-band navigation radar is mounted high on the ship, scanning the sea surface. By analyzing the backscattered electromagnetic waves from the waves, it retrieves a two-dimensional wave spectrum over a wide area of ​​several kilometers around the ship. This two-dimensional wave spectrum provides the macroscopic statistical characteristics of the waves, including wave frequency and propagation direction. The lidar system is also mounted on the bow, performing high-frequency scanning of the near-field area several hundred meters ahead of the bow. By measuring the flight time of the laser pulses, it directly obtains an instantaneous three-dimensional waveform profile within this area with centimeter-level accuracy, i.e., the sea surface elevation function.

[0061] The environmental perception module also includes a ship status data acquisition unit, which consists of an inertial measurement unit (IMU), a global positioning system (GPS), and a propulsion system sensor array. The IMU measures the ship's six degrees of freedom motion attitude in real time, outputting dynamic information including angular displacement (including pitch, roll, and heave), angular velocity, and linear acceleration. The GPS provides the ship's absolute geographical location, ground speed, and heading angle. The propulsion system sensor array directly monitors the operating status of the ship's power plant, specifically including a main engine speed sensor, torque sensor, and fuel flow meter, used to acquire the main engine's real-time speed, output torque, and instantaneous fuel consumption rate. For ships equipped with adjustable-pitch propellers, this ship status data acquisition unit also includes a pitch sensor to obtain the propeller's real-time pitch.

[0062] In addition, the environmental perception module is equipped with an external environmental data acquisition unit, such as an anemometer, to obtain the relative wind speed and relative wind direction angle encountered by the ship during navigation, providing data support for subsequent separation of wind-induced effects and accurate assessment of wave-induced response.

[0063] To integrate the aforementioned multi-source heterogeneous data, the environmental perception module includes a data synchronization and preprocessing unit. This unit receives the raw data streams from the sensors, assigns a unified, high-precision timestamp to all input data to ensure accurate temporal alignment of different physical quantities, and performs filtering on the raw data, such as using median filtering or low-pass filtering algorithms, to remove random noise and high-frequency disturbances from the sensors. All data after timestamp synchronization and filtering is then uniformly converted into a standardized data format, forming structured data frames, which are continuously transmitted to the state prediction module as the final output of the environmental perception module.

[0064] The state prediction module is used to deeply fuse multi-source and heterogeneous data provided by the environmental perception module to construct a deterministic wave field model describing the local sea area in front of the bow. Based on this model, it can make real-time and high-precision predictions of the wave sequence that the ship will encounter in the short time domain (e.g., 10 to 30 seconds) and the multi-dimensional responses that may be triggered.

[0065] The state prediction module includes a wave field reconstruction and forward prediction unit. The goal of the wave field reconstruction and forward prediction unit is to generate a deterministic waveform sequence prediction model for feedforward control. The specific implementation method is as follows: the local wave field is represented as the result of the linear superposition of multiple regular waves with different frequencies, directions and phases. At a specific time, the state of the wave field is represented by a state vector.

[0066] To accurately estimate the state vector, the wavefield reconstruction and forward prediction unit employ a Kalman filter framework to fuse multi-source data. The wide-area wave spectrum provided by the X-band radar is used as statistical prior information, providing the initial distribution of amplitude, frequency, and direction in the state vector. Based on linear wave theory, the evolution of the state over time is primarily characterized by a linear phase progression, forming the Kalman-filtered state transition model. Near-field high-precision sea surface elevation measurements provided by the lidar system serve as observation input, used to correct and update the state vector estimate at each time step. Through this fusion mechanism, the model retains both the statistical characteristics of the wide-area sea state and incorporates the deterministic details of the near-field waveform.

[0067] After obtaining the optimized current wavefield state vector, the wavefield reconstruction and forward prediction unit can then predict the precise waveform sequence η(x,y,t) of any position (x,y) that the hull will encounter within the future time domain T. k Forward computation is performed using +τ)(where 0 < τ ≤ T):

[0068]

[0069] Where, η(x,y,t) k+τ) represents the precise waveform (sea surface elevation) that the ship will encounter at any position (x,y) at the future time τ; i is the index of the wave component; N is the total number of wave components; a i Its amplitude; ω i Its natural frequency; θ i Its propagation direction; (x,y) represents the coordinates of any position in the ship's coordinate system; t k For the current moment; φ i,k For time t k The phase; τ represents the phase of the current time t. k Future prediction time; k i Let be the wave number of the i-th wave component, whose value is related to the frequency ω. i It satisfies the dispersion relation. The deterministic waveform sequence predicted by this method is the direct input for all subsequent ship response predictions.

[0070] The state prediction module also includes a multi-dimensional ship response prediction unit, which receives deterministic waveform sequences and, in conjunction with a pre-installed ship hydrodynamic performance database in the system, predicts in real time the ship's state in three dimensions: energy efficiency, structural load, and cargo safety within the next T seconds.

[0071] In terms of energy efficiency state prediction, this multi-dimensional ship response prediction unit calculates the additional drag of the ship in waves based on the predicted waveform sequence and the ship's response amplitude operator (RAO) or other fast time-domain models. Simultaneously, considering the changes in propeller inflow velocity and immersion depth caused by the predicted hull motion (especially pitch and heave), this multi-dimensional ship response prediction unit assesses the fluctuations in the propeller's instantaneous thrust and torque coefficient in waves, thereby predicting the dynamic changes in propulsion efficiency. Combining total drag (the sum of still water drag and additional drag) with propulsion efficiency, the unit ultimately predicts the instantaneous main engine power requirement and fuel consumption rate needed to maintain the current average speed.

[0072] In terms of structural load prediction, this multi-dimensional ship response prediction unit utilizes a pre-calculated total longitudinal bending moment response amplitude operator of the hull, combined with the wave component information in the predicted waveform sequence, to predict the total longitudinal bending moment at key locations such as the midship section, including wave encounter frequency and encounter angle. Similarly, it can also predict other key structural loads such as the total longitudinal shear force of the hull.

[0073] In terms of cargo safety status prediction, this multi-dimensional ship response prediction unit first calculates the motion state of a specific cargo position on the ship based on the predicted six-degree-of-freedom motion response of the hull and the rigid body kinematic equations. Then, the unit derives the vertical and lateral accelerations at that position and synthesizes them into a comprehensive acceleration index to quantify the impact intensity and potential risks experienced by the cargo at that moment.

[0074] Ultimately, the state prediction module combines all predicted state time series within the next T seconds, including predicted fuel consumption rate, predicted total longitudinal bending moment, and predicted cargo position comprehensive acceleration, into a structured predicted state dataset, and outputs it to the intelligent decision-making module, providing a deterministic and forward-looking basis for its multi-objective optimization decision-making.

[0075] See attached document Figure 3 The intelligent decision-making module receives multi-dimensional state prediction information about the future short time domain from the state prediction module, and makes dynamic optimization decisions based on this information, ultimately generating the optimal and coordinated control commands.

[0076] This intelligent decision-making module employs a model predictive control (MPC) framework. In each control cycle, the module solves a multi-objective optimization problem over the future prediction time domain T, aiming to find a control input sequence that minimizes the overall cost function. This control input sequence specifically includes adjustments to the main engine speed and the rudder angle.

[0077] The energy efficiency cost is directly related to the instantaneous fuel consumption rate predicted by the state prediction module, and is used to reduce the average fuel consumption during navigation.

[0078] The structural safety cost is a nonlinear penalty function relating to the predicted total longitudinal bending moment. When the predicted total longitudinal bending moment approaches a preset allowable bending moment threshold, the cost function value will increase sharply to penalize any control strategies that may jeopardize the structural safety of the hull.

[0079] The cargo safety cost is a nonlinear penalty function of the predicted combined acceleration of the cargo position, and its form is similar to that of the structural safety cost. When the predicted acceleration value approaches the safe acceleration threshold that the cargo can withstand, the cost function value will also increase sharply to avoid violent hull movements that may cause cargo displacement or damage.

[0080] The intelligent decision-making module includes a risk assessment and weight adaptive unit, with a weight factor w. f ,w s ,w c The risk assessment and weighted adaptive unit is dynamically adjusted based on the predicted risk output by the state prediction module. This unit is first based on the predicted structural load M. BM A comprehensive risk level index R is calculated by taking the peak values ​​of A(τ) and cargo acceleration A(τ) over the entire prediction time domain T. level Subsequently, the weighting factor is set as a function of the risk level indicator, for example, w. s =f s (R level ) and w c =f c (R level), where f s (·) and f c (·) is a non-decreasing function, which enables the entire control system to seamlessly switch adaptively between different control modes.

[0081] In predicting low risk (R) level When the value is small, the weight w s and w c Approaching 0, making w f When the value is close to 1, the system automatically enters the energy efficiency priority mode, at which point the optimization objective of the cost function J is mainly focused on minimizing fuel consumption.

[0082] When a dangerous wave group is predicted to be encountered, the predicted peak load or acceleration increases (R). level When the value increases, the weight w s and w c The value of will increase significantly, and the system will automatically switch to the safety priority mode. At this time, the optimization objective of the cost function J will be to prioritize avoiding structural or cargo risks, even if the control actions taken will temporarily sacrifice some energy efficiency.

[0083] The intelligent decision-making module also includes an optimal control sequence solving unit. This unit employs numerical optimization algorithms, such as Sequential Quadratic Programming (SQP) or interior-point methods, to solve the multi-objective dynamic cost function. The solution process adheres to physical constraints, including:

[0084] The system defines the maximum and minimum rotational speeds of the main engine, the rate of increase and decrease of main engine power, the maximum usable rudder angle of the servo motor, and the maximum steering rate limit. By solving this constrained optimization problem, the optimal control sequence solving unit calculates the optimal control sequence that minimizes the comprehensive cost function J within the future predicted time domain T. This optimal control sequence includes refined strategies for coordinated control of the main engine and servo motor, such as "pre-acceleration before waves," "pre-power reduction before peaks," or "drag reduction micro-steering." Finally, the intelligent decision-making module outputs the calculated optimal control sequence to the collaborative execution module.

[0085] See attached document Figure 4 The collaborative execution module is used to transform the optimal control commands generated by the intelligent decision-making module into specific, physically executable operations on the ship's propulsion and maneuvering systems.

[0086] The collaborative execution module includes an instruction parsing and distribution unit, which receives the complete optimal control sequence from the intelligent decision-making module in each control cycle and extracts only the first element of the sequence.

[0087] The collaborative execution module includes a propulsion system control interface unit, which receives the main engine speed adjustment command Δn. e (0) is then converted into a standard signal for the corresponding underlying propulsion system controller. For ships equipped with fixed-pitch propellers, this signal is sent to the main engine governor, which precisely adjusts the fuel injection quantity to change the main engine speed. For ships equipped with adjustable-pitch propellers, this main engine speed adjustment command can be further interpreted as a coordinated adjustment of the main engine speed and propeller pitch to ensure that the propulsion system operates at its highest efficiency point while meeting thrust requirements.

[0088] The collaborative execution module also includes a control system interface unit, which receives rudder angle adjustment commands and sends them to the execution unit of the autopilot system. The execution unit of the autopilot system then drives the servo motor to achieve precise and rapid deflection of the rudder angle.

[0089] This collaborative execution module implements control strategies driven by accurate prediction and intelligent decision-making, including:

[0090] Pre-acceleration before waves: When the state prediction module predicts that the bow will encounter a large wave that will generate significant additional drag, the intelligent decision-making module will generate a small positive power adjustment command. Based on this, the cooperative execution module will increase the main engine output power in advance and smoothly to compensate for the upcoming wave-induced drag, thereby avoiding a sharp drop in ship speed and subsequent power overcompensation, achieving the effect of smooth energy consumption and stable speed.

[0091] Peak load reduction: When it is predicted that the ship will soon pass through a wave peak, which may increase the risk of propeller idling or cause a sudden decrease in load, the cooperative execution module will execute a momentary, small-amplitude power reduction command. This proactively adapts to the upcoming load change, saving energy and improving the operational safety of the propulsion system.

[0092] Rolling drag reduction micro-steering: The rudder angle adjustment command executed by the cooperative execution module will perform a small (e.g., less than one degree) and precise rudder angle deflection at a specific wave encounter angle. This is used to actively generate a roll suppression torque to counteract the roll caused by the waves, or to fine-tune the relative attitude of the hull and the waves in order to reduce wave-induced additional drag. These fine-tuning will not significantly deviate from the predetermined course.

[0093] The performance feedback module is used to evaluate the actual operational performance of the aforementioned micro-level collaborative control, and to digitize and model this performance data, ultimately feeding it back to the top-level macro-level route planning system to achieve adaptive adjustment of macro-level strategic decisions to micro-level tactical capabilities.

[0094] The performance feedback module contains a dynamic performance evaluation unit. This unit continuously receives and records real-time sea state parameters (including significant wave height, main wave period, and encounter angle between the ship and the waves) from the environmental perception module, as well as the actual instantaneous fuel consumption rate from the propulsion system sensors. Simultaneously, this unit invokes a pre-installed ship baseline performance model, which describes the benchmark fuel consumption rate for the ship under the same sea state parameters when employing conventional control strategies (e.g., constant speed or constant power).

[0095] This dynamic performance evaluation unit calculates the energy efficiency gain factor η, which quantifies the energy-saving potential of the microcontroller system, by comparing actual consumption with baseline consumption. gain The energy efficiency gain factor is calculated as follows:

[0096]

[0097] Where, η gain Energy efficiency gain factor; (H) s ,T p ,χ) represents the sea state parameter; FC actual Actual fuel consumption rate; FC baseline H is the baseline fuel consumption rate. s For significant wave height; T p η is the main wave period; χ is the encounter angle between the ship and the wave. gain The value reflects the percentage of fuel savings achieved by the collaborative control strategy of this invention compared to conventional strategies.

[0098] The performance feedback module also includes a dynamic performance graph construction unit, which is responsible for collecting numerous data points (η) generated by the dynamic performance evaluation unit. gain H s ,T p ,χ), and use this data to construct and continuously update a multi-dimensional "dynamic performance map" E(H) online. s ,T p The dynamic performance graph E (χ) can be mathematically a multidimensional look-up table or a surrogate model fitted by machine learning algorithms (such as Gaussian process regression or neural networks). This dynamic performance graph represents the actual, achievable energy efficiency gain of the intelligent control system on board the ship under specific sea conditions.

[0099] The performance feedback module includes a macro-level route cost function correction unit, which will construct the dynamic performance graph E(H). s ,T pχ) is used as a correction parameter and provided to the weather navigation system. When performing global route optimization, the objective of a conventional weather navigation system is to minimize the total fuel consumption (TFC), and its cost function is usually expressed as:

[0100]

[0101] Where, minTFC is the objective function; TFC is the total fuel consumption; ∫ Route This indicates that points are accumulated along the entire planned route; FC baseline (V g H s ,T p ,χ) represents the baseline fuel consumption rate; H s For significant wave height; T p The main wave period; χ is the encounter angle between the ship and the wave; ds is the course element; V g This is the ground speed for this segment of the journey.

[0102] In this embodiment, the cost function correction unit modifies the cost function by introducing a dynamic performance graph E, thereby obtaining an adaptive cost function TFC. new :

[0103]

[0104] Among them, E(H s ,T p ,χ) represents the dynamic performance spectrum; minTFC new Minimize the corrected total fuel consumption; TFC new Corrected total fuel consumption.

[0105] By optimizing this modified cost function TFC new When planning routes, the macro-meteorological navigation system shifts its decision-making basis from "which path has smaller winds and waves" to "on which path can the ship's intelligent control system achieve the greatest energy-saving potential," forming a complete adaptive optimization closed loop.

[0106] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An energy efficiency optimization system for ultra-large container ships based on intelligent navigation control, characterized in that, include: The environmental perception module is used to collect multi-source data on the ship's navigation environment and its own status in real time; The state prediction module is used to receive and fuse the multi-source data to construct a local deterministic wavefield model, and to predict the waveform sequence in the short time domain and the multi-dimensional ship response caused by the waveform sequence based on the local deterministic wavefield model, so as to obtain multi-dimensional state prediction information. The intelligent decision-making module is used to receive the multi-dimensional state prediction information, use model predictive control methods to solve the optimization problem of minimizing the multi-objective dynamic cost function, and generate the optimal cooperative control command sequence for the host and servo motor. The collaborative execution module is used to parse the optimal collaborative control command sequence and convert it into specific control signals for the ship's propulsion system and maneuvering system to coordinate actions; The performance feedback module is used to evaluate the actual operating performance of the collaborative execution module under different sea conditions, construct and update a dynamic performance map describing the energy-saving potential, and feed the dynamic performance map back to the ship's macro-meteorological navigation system to correct the cost function of global route optimization.

2. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 1, characterized in that, The environment sensing module includes: The local sea state data acquisition unit integrates an X-band navigation radar for inverting wide-area two-dimensional wave spectra and a lidar system for directly acquiring near-field instantaneous three-dimensional waveform profiles. The ship status data acquisition unit consists of an inertial measurement unit, a global positioning system, and a propulsion system sensor group used to acquire the real-time speed, output torque, and instantaneous fuel consumption rate of the main engine; The external environment data acquisition unit includes an anemometer for acquiring relative wind speed and relative wind direction angle; The data synchronization and preprocessing unit is used to assign unified timestamps and perform filtering processing on the multi-source data collected by the local sea state data acquisition unit, the ship status data acquisition unit and the external environment data acquisition unit to form standardized structured data frames.

3. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 1, characterized in that, The state prediction module includes: The wave field reconstruction and forward prediction unit is used to fuse the multi-source data using a Kalman filter framework. Specifically, it uses the wide-area wave spectrum provided by the X-band navigation radar as statistical prior information and the near-field high-precision sea surface elevation measurement provided by the lidar system as observation input to estimate and update the state vector of the local wave field and perform forward calculation on the waveform sequence in the future prediction time domain. The multi-dimensional ship response prediction unit is used to receive waveform sequences and, in conjunction with a pre-set ship hydrodynamic performance database, predict multi-dimensional response states in real time.

4. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 3, characterized in that, The real-time prediction of multi-dimensional response status specifically includes: Instantaneous fuel consumption rate related to energy efficiency; Total longitudinal bending moment related to structural loads; Overall acceleration of cargo position.

5. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 1, characterized in that, The intelligent decision-making module is specifically used for: Based on the multi-dimensional state prediction information, the multi-objective dynamic cost function is constructed. In the prediction time domain, the multi-objective dynamic cost function performs a weighted summation of multiple cost items representing operational energy efficiency, structural safety and cargo safety. The energy efficiency cost item is determined based on the predicted instantaneous fuel consumption rate, the structural safety cost item is determined based on the predicted total longitudinal bending moment, and the cargo safety cost item is determined based on the predicted comprehensive acceleration of cargo position.

6. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 5, characterized in that, The intelligent decision-making module includes a risk assessment and weight adaptation unit, specifically used for: The comprehensive risk level index is calculated based on the peak value of the combined acceleration of the predicted structural load and cargo position throughout the entire prediction time domain. The dynamic weighting factor is dynamically adjusted based on the comprehensive risk level index, and the dynamic adjustment steps include: When the comprehensive risk level index is low, the weighting factor of energy efficiency cost is increased, while the weighting factors of structural safety cost and cargo safety cost are decreased, in order to operate in an energy efficiency priority mode. When the comprehensive risk level index increases, the weighting factors of structural safety cost and cargo safety cost are increased to automatically switch to a safety-first mode.

7. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 6, characterized in that, The intelligent decision-making module also includes an optimal control sequence solving unit, specifically used for: A numerical optimization algorithm is used to solve the multi-objective dynamic cost function after the weights are adjusted by the risk assessment and weight adaptation unit. The solution process follows physical constraints, which include: the maximum and minimum rotational speeds of the main engine, the power increase / decrease rate limit of the main engine, the maximum usable rudder angle of the rudder, and the maximum rudder turning rate limit.

8. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 1, characterized in that, The collaborative actions performed by the collaborative execution module specifically include: When the intelligent decision-making module receives a positive power adjustment command based on the prediction that the bow will encounter a large wave that will generate significant additional drag, the main engine output power is increased in advance to compensate for the wave-induced drag. When the intelligent decision-making module receives a power reduction command based on a prediction that the ship is about to pass through a wave crest, which may increase the risk of propeller idling or cause a sudden decrease in load, it will execute an instantaneous and small-amplitude power reduction to proactively adapt to load changes. When the intelligent decision-making module receives a rudder angle adjustment command generated to reduce resistance at a specific wave encounter angle, the rudder angle is deflected to actively generate a roll suppression torque to counteract the roll caused by the waves, or to adjust the relative attitude of the hull and the waves to reduce wave-induced additional resistance. The rudder angle deflection will not cause the ship to deviate significantly from the predetermined course.

9. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 1, characterized in that, The performance feedback module includes: The dynamic performance evaluation unit is used to receive the instantaneous fuel consumption rate of the ship's navigation environment and propulsion system sensors from the environmental perception module, and call the preset ship baseline performance model to obtain the benchmark fuel consumption rate under the same ship navigation environment, and calculate the energy efficiency gain factor by comparison. The dynamic performance map construction unit is used to collect multiple data points generated by the dynamic performance evaluation unit under different ship navigation environments in order to construct a multi-dimensional dynamic performance map that is updated online and describes the energy-saving potential. The macro-route cost function correction unit is used to provide the multi-dimensional dynamic performance map as a correction parameter to the macro-meteorological navigation system in order to correct the cost function of global route optimization.

10. The energy efficiency optimization system for ultra-large container ships based on intelligent navigation control according to claim 9, characterized in that, The pre-set ship baseline performance model is a performance reference model that describes the benchmark fuel consumption rate of a ship under specific combinations of speed, draft, wind conditions, and sea conditions, established in advance through ship tank tests, numerical simulation calculations, or actual ship navigation test data.

Images