High-efficiency ship automatic control system

Abstract

The invention provides a high-efficiency ship automatic control system, and the system comprises a comprehensive control platform which carries out the unified integration of subsystems, sets a target function, and carries out the data sharing, collaborative optimization and global control among the subsystems. The intelligent sensing network module is used for collecting historical data and real-time data; the self-learning control module is used for constructing a control strategy generation model; the ship digital model module is used for constructing a digital model for simulating ship navigation so as to test different ship control strategies; the decision execution module is used for generating and executing a control instruction; and the safety monitoring module detects abnormal behaviors of the ship and gives an alarm. According to the invention, a plurality of subsystem modules can be integrated in a unified manner, data sharing, collaborative optimization and global control are realized, an optimal control strategy can be provided, the cost and risk of actual navigation are reduced, and the control precision, response speed and operation efficiency of a ship are improved.

Description

Technical Field

[0001] The present invention relates to the field of ship control technology, in particular to a high-efficiency ship automatic control system. Background Art

[0002] Ship automatic control system refers to a system that uses computer control technology and electronic technology to achieve automated management of ship navigation under the premise of ensuring the safety of ship operation. It mainly includes power measurement and control system, steering gear system, sail and rudder system and electronic navigation system, etc. Its purpose is to improve the operation efficiency, safety and economy of the ship and reduce the complexity and errors of human operation.

[0003] Although the existing ship automatic control system has made great progress in technology, there is a lack of unified integration between different control systems (such as autopilot, dynamic positioning, and collision avoidance systems), resulting in information islands and operational complexity. In addition, most existing systems are rule-driven and lack self-learning capabilities, resulting in limited intelligence levels, making it difficult to adapt to dynamically changing navigation environments and unable to achieve true autonomous navigation. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a high-efficiency ship automatic control system, which can not only integrate multiple subsystem modules to achieve data sharing, collaborative optimization and global control, but also provide optimal control strategies, reduce the actual navigation costs and risks, and improve the ship's control accuracy, response speed and operation efficiency.

[0005] To achieve the above-mentioned object, the present invention provides the following scheme: a high-efficiency ship automatic control system, comprising an integrated control platform and an intelligent sensor network module, a self-learning control module, a ship digital model module, a decision execution module and a safety monitoring module connected to the integrated control platform;

[0006] The integrated control platform is used to integrate the heading control subsystem, speed control subsystem, dynamic positioning subsystem, collision avoidance subsystem and energy efficiency management subsystem, and set the objective function to carry out data sharing, collaborative optimization and global control among the subsystems;

[0007] The intelligent sensor network module is used to use multiple sensors to perform data collection and data fusion to obtain raw data, perform data preprocessing on the raw data to obtain multimodal data, and use edge computing devices to perform real-time data collection;

[0008] The self-learning control module is used to select a deep neural network and a long short-term memory network based on a multi-task learning framework, and then use the objective function to perform model training, generate a control strategy generation model, and update the control strategy generation model according to real-time data;

[0009] The ship digital model module is used to construct a physical model and an environmental model of the ship, obtain a digital model simulating the navigation of the ship, and use the digital model to test different ship control strategies;

[0010] The decision execution module is used to convert the output of the control strategy generation model into a control instruction, and send the control instruction to each execution mechanism of the ship;

[0011] The safety monitoring module is used to detect abnormal behavior of the ship and to perform graded alarms and records according to the detection results.

[0012] Optionally, the integrated control platform includes:

[0013] Data linkage unit, which is used to design data bus and communication protocol for data sharing between subsystems, standardize the data of each subsystem, synchronize subsystem data through timestamp mechanism, and achieve data consistency;

[0014] A subsystem linkage unit is used to adjust the speed output power through the heading control target to realize the linkage between the heading control subsystem and the speed control subsystem, to realize the linkage between the dynamic positioning subsystem and the collision avoidance subsystem by introducing collision avoidance logic into the dynamic positioning subsystem, to adjust the speed and propeller power by monitoring the fuel consumption and emission data, and to realize the global optimization of the energy efficiency management subsystem;

[0015] The collaborative optimization unit is used to define the objective function with fuel consumption, safety cost and navigation efficiency as the optimization targets and perform global optimization control;

[0016] The human-computer interaction unit is used to design the user interface, display the ship's operating status and environmental information in real time, and present the data intuitively through dashboards, heat maps or navigation path maps.

[0017] Optionally, the objective function definition process includes: setting the minimum fuel consumption and the balance point between fuel efficiency and navigation time as optimization targets, and setting the collision avoidance distance and heading deviation as safety constraints, combining the optimization target and the safety constraints to obtain the objective function, and introducing a genetic algorithm or a particle swarm optimization algorithm into the objective function to calculate the multi-objective optimization problem; the formula of the objective function is:

[0018] F=ω1·C f +ω2·S c +ω3·E t

[0019] Among them, F is the objective function value, C f is the fuel consumption, Sc is the security cost function, E t is the navigation efficiency cost function, ω1, ω2, ω3 are all weight coefficients, ω1+ω2+ω3=1.

[0020] Optionally, the calculation formula for the fuel consumption is:

[0021]

[0022] Wherein, P(t) is the output power of the ship's main engine at time t, SFC is the specific fuel consumption rate of the main engine, and T is the actual sailing time;

[0023] The calculation formula of the security cost function is:

[0024]

[0025] Where N is the total number of obstacles during navigation, d i is the distance between the ship and the i-th obstacle, Δθ i is the deviation between the current heading of the ship and the planned heading for collision avoidance, k i is the weight coefficient used to balance the importance of collision avoidance distance and heading deviation;

[0026] The calculation formula of the navigation efficiency cost function is:

[0027]

[0028] Among them, T is the actual navigation time, T optimal is the ideal sailing time, C f is the actual fuel consumption, C f·optimal is the ideal fuel consumption, α is the time cost weight coefficient used to represent the importance of time efficiency, and β is the fuel cost weight coefficient used to represent the importance of fuel efficiency.

[0029] Optionally, the intelligent sensor network module includes:

[0030] Historical data acquisition unit, used to deploy lidar, ultrasonic sensors, inertial navigation systems, meteorological sensors, GPS, radar and AIS to perform environmental perception and ship status monitoring and obtain multi-source sensor data;

[0031] A data fusion unit is used to synchronize the multi-source sensor data with the timestamp, and to align the multi-source sensor data with the unified coordinate system, and then combine the distance data of the laser radar and ultrasonic sensor to generate obstacle information to complete low-level fusion, and combine the GPS, radar and AIS data to generate a ship distribution map to complete high-level fusion to obtain the original data;

[0032] A real-time data collection unit is used to deploy multiple edge computing devices in different areas of the ship in a combination of wired and wireless methods to collect real-time data;

[0033] The data preprocessing unit is used to perform noise filtering, anomaly detection and data standardization on the raw data and real-time data to obtain multimodal data.

[0034] Optionally, the self-learning control module includes:

[0035] A data collection unit, used to collect the multimodal data, environmental data and ship operation data, and perform data cleaning, normalization and feature extraction on the collected data to obtain a comprehensive data set;

[0036] A model building unit, used to process nonlinear data in the comprehensive data set using a deep neural network and process time series data in the comprehensive data set using a long short-term memory network based on a multi-task learning framework to obtain an initial model;

[0037] A model training unit, used to train the initial model in a supervised learning manner using the multi-source sensor data in combination with the objective function to obtain a control strategy generation model;

[0038] The model updating unit is used to perform incremental learning of the control strategy generation model using real-time data to achieve parameter optimization.

[0039] Optionally, the ship digital model module includes:

[0040] A digital model building unit is used to build a physical model of the ship, introduce fluid mechanics into the physical model, simulate the dynamic characteristics of the ship under different speeds and sea conditions, build an external environment model, and generate a simulated navigation scene;

[0041] A strategy optimization unit is used to perform scenario simulation in combination with the ship physical model and the simulated navigation model to test different ship control strategies, and optimize the control strategy generation model according to the test results.

[0042] Optionally, the decision execution module includes:

[0043] An instruction generation module is used to convert the output of the control strategy generation model into a control instruction with a unified format, and send the control instruction to the corresponding actuator through a communication network and a redundant communication mechanism; the control instruction includes a rudder angle adjustment instruction, a thruster power distribution instruction, a speed adjustment instruction and an auxiliary equipment control instruction;

[0044] The execution feedback module is used to adjust the output of the control strategy generation model by feeding back the execution result of the control instruction to the control strategy generation model.

[0045] Optionally, the safety monitoring module includes:

[0046] An anomaly detection unit is used to monitor ship operation parameters, environmental parameters and equipment status in real time using a variety of sensors and edge computing devices, and perform anomaly detection according to a preset anomaly range to obtain an anomaly detection result, and then classify the anomaly detection result into low-risk anomaly, medium-risk anomaly and high-risk anomaly according to the severity of the anomaly detection result;

[0047] An alarm unit is used to preset alarm trigger conditions and perform graded alarms according to the classification of the abnormality detection results; the alarm trigger conditions include the distance between the ship and the obstacle being lower than a safety threshold, the main engine temperature or pressure exceeding a safety range, and the ship deviating from the scheduled route; the graded alarms include a level one alarm for the high-risk abnormalities, a level two alarm for the medium-risk abnormalities, and a level three alarm for the low-risk abnormalities.

[0048] The present invention discloses the following technical effects by providing a high-efficiency ship automatic control system:

[0049] 1. The present invention can integrate system modules such as heading control, speed control, dynamic positioning, collision avoidance system, and energy efficiency management by setting up an integrated control platform, thus realizing data sharing, subsystem collaborative optimization, and global control. It makes fuel efficiency, safety, and navigation efficiency reach global optimization, reduces operating costs, improves navigation stability under complex sea conditions, and lays a technical foundation for ship automation and intelligence through unified integration and intelligent optimization algorithms.

[0050] 2. The present invention realizes self-learning control strategy and utilizes control strategy generation model to process linear and nonlinear relationships, so that the relationship between fuel consumption and speed, propeller power, and the impact of environmental changes on control strategy during navigation can be deeply processed to obtain the optimal control strategy, improve the ship's control accuracy, response speed and operating efficiency, enable the system to respond quickly in complex sea conditions and dynamic environments, and achieve efficient and safe navigation.

[0051] 3. Through the application of digital models, the present invention can simulate various working conditions (such as severe sea conditions and emergency collision avoidance) through virtual experiments, verify the effectiveness of control strategies, optimize control parameters and collision avoidance strategies, and reduce the cost and risk of actual navigation.

[0052] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0054] Figure 1 A schematic diagram of a system architecture provided by an embodiment of the present invention;

[0055] Figure 2 A schematic diagram of the architecture of the integrated control platform provided by an embodiment of the present invention;

[0056] Figure 3 A schematic diagram of a flow chart for generating a control strategy provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] like Figure 1 As shown, the present invention provides a high-efficiency ship automatic control system, including an integrated control platform and an intelligent sensor network module, a self-learning control module, a ship digital model module, a decision execution module and a safety monitoring module connected to the integrated control platform.

[0060] 1. Integrated control platform

[0061] like Figure 2 As shown, it is used to integrate the heading control subsystem, speed control subsystem, dynamic positioning subsystem, collision avoidance subsystem and energy efficiency management subsystem, and set the objective function to carry out data sharing, collaborative optimization and global control among subsystems. Optionally, the integrated control platform includes:

[0062] 1.1 Data linkage unit

[0063] Design data bus and communication protocol to share data between subsystems, standardize data of each subsystem, synchronize subsystem data through timestamp mechanism, and achieve data consistency;

[0064] 1.2 Subsystem linkage unit

[0065] Heading and speed linkage: According to the heading control target, the output power of the speed control module is dynamically adjusted to ensure that the ship maintains a stable navigation state under complex sea conditions, and the heading control subsystem and the speed control subsystem are linked.

[0066] Dynamic positioning and collision avoidance linkage: Introduce collision avoidance logic in the dynamic positioning system. When an obstacle is detected, the output of the thruster and the steering gear is adjusted first to avoid collision, so as to realize the linkage between the dynamic positioning subsystem and the collision avoidance subsystem.

[0067] Energy efficiency management and global optimization: The energy efficiency management module monitors fuel consumption and emission data in real time, and feeds back optimization suggestions to the speed control and dynamic positioning modules, dynamically adjusts the speed and propeller power, reduces fuel consumption, and realizes global optimization of the energy efficiency management subsystem;

[0068] 1.3 Collaborative Optimization Unit

[0069] It is used to define the objective function with fuel consumption, safety cost and navigation efficiency as optimization targets, and perform global optimization control.

[0070] The definition process of the objective function includes:

[0071] Minimize fuel consumption: Take fuel consumption as the main optimization goal.

[0072] Maximize navigation safety: with collision avoidance distance, heading deviation, etc. as constraints.

[0073] Balancing navigation efficiency: Finding the best balance between fuel efficiency and navigation time is the main optimization goal.

[0074] Combining the optimization goal and the safety constraint condition, an objective function is obtained, and a genetic algorithm (using a genetic algorithm to solve multi-objective optimization problems and generate a global optimal control strategy) or a particle swarm optimization algorithm is introduced into the objective function to calculate the multi-objective optimization problem; the formula of the objective function is:

[0075] F=ω1·C f +ω2·S c +ω3·E t

[0076] Among them, F is the objective function value (need to be minimized), C f is the fuel consumption (need to be minimized), S c is the security cost function (need to be minimized), E tis the navigation efficiency cost function (needs to be minimized), ω1, ω2, ω3 are all weight coefficients, which are used to balance the importance of fuel consumption, safety and navigation efficiency, and can be adjusted according to actual needs, ω1+ω2+ω3=1.

[0077] The calculation formula for the fuel consumption is:

[0078]

[0079] Where P(t) is the output power of the ship's main engine at time t (kW), SFC is the main engine's specific fuel consumption (g / kWh) (usually a load-related nonlinear function), and T is the actual sailing time (h).

[0080] Fuel consumption is the product of main engine power and specific fuel consumption, which is reduced by optimizing speed and propeller efficiency to reduce power demand and thus reduce fuel consumption.

[0081] The calculation formula of the security cost function is:

[0082]

[0083] Where N is the total number of obstacles during navigation, d i is the distance between the ship and the ith obstacle (meters), Δθ i is the deviation (degrees) between the current course of the ship and the planned course for collision avoidance, k i is a weight coefficient used to balance the importance of collision avoidance distance and heading deviation (determined according to the ratio of collision avoidance distance to heading deviation);

[0084] When the ship is close to an obstacle, the value of the safety cost function will increase rapidly, and the system will prioritize adjusting the course or speed to increase the collision avoidance distance; the course deviation control ensures that the ship can quickly return to the established route after avoiding a collision to avoid excessive deviation from the path.

[0085] The calculation formula of the navigation efficiency cost function is:

[0086]

[0087] Where, T is the actual sailing time (hours), T optimal is the ideal sailing time, C f is the actual fuel consumption (g), C f·optimal is the ideal fuel consumption, α is the time cost weight coefficient used to represent the importance of time efficiency, and β is the fuel cost weight coefficient used to represent the importance of fuel efficiency.

[0088] The navigation efficiency cost function measures the navigation efficiency by the ratio of actual navigation time to ideal navigation time, and the ratio of actual fuel consumption to ideal fuel consumption; the weights of α and β can be dynamically adjusted according to mission requirements. For example, α can be increased in emergency missions, while β can be increased when energy saving is prioritized.

[0089] 1.4 Human-computer interaction unit

[0090] It is used to design user interfaces, display the ship's operating status (such as speed, heading, fuel consumption, rudder angle, propeller power, etc.) and environmental information (such as wind speed, wave height, obstacle location, etc.) in real time, and present the data intuitively through dashboards, heat maps or navigation path maps, so that the crew can quickly understand it. In addition, it can provide a variety of interactive methods such as touch screens, physical buttons, and voice control to meet different operating habits.

[0091] 2. Intelligent sensor network module

[0092] It is used to collect and fuse data using multiple sensors to obtain raw data, preprocess the raw data to obtain multimodal data, and use edge computing devices to collect real-time data. The intelligent sensor network module includes:

[0093] 2.1 Historical data collection unit

[0094] It is used to deploy lidar, ultrasonic sensors, inertial navigation systems, meteorological sensors, GPS, radar and AIS to perform environmental perception and ship status monitoring and obtain multi-source sensor data. Deploying multiple sensors can achieve data redundancy and ensure that the system can still operate normally when a single sensor fails.

[0095] Laser Radar (LiDAR): Used to accurately measure the distance and shape of surrounding obstacles and generate high-resolution three-dimensional environmental maps.

[0096] Ultrasonic sensors: used for close-range obstacle detection, especially providing highly accurate distance information in narrow waters or when docking.

[0097] Inertial Navigation System (INS): monitors the ship's attitude, heading and motion status in real time through accelerometers and gyroscopes.

[0098] Meteorological sensors: real-time monitoring of environmental parameters such as wind speed, wind direction, air pressure, temperature and humidity to provide support for navigation decisions.

[0099] GPS / Beidou Navigation System: Provides high-precision ship position and speed information.

[0100] Radar and AIS (Automatic Identification System): used to monitor the dynamic information of surrounding ships and assist in collision avoidance and path planning.

[0101] 2.2 Data Fusion Unit

[0102] It is used to synchronize the multi-source sensor data using timestamps to ensure the consistency of data at the same time point. It also uses a unified coordinate system to spatially align the multi-source sensor data, and then combines the distance data of the laser radar and ultrasonic sensor (direct fusion of raw data) to generate obstacle information, complete low-level fusion, and combine GPS, radar and AIS data (fusion of sensor processing results) to generate a ship distribution map, complete high-level fusion, and obtain raw data.

[0103] 2.3 Real-time data acquisition unit

[0104] It is used to deploy multiple high-performance edge computing devices (such as embedded GPU or FPGA) in different areas of the ship (such as the cockpit, engine room, and deck) in a combination of wired and wireless methods to collect real-time data.

[0105] 2.4 Data Preprocessing Unit

[0106] It is used to perform noise filtering, anomaly detection and data standardization on the raw data and real-time data to obtain multimodal data.

[0107] 2.41 Noise Filtering

[0108] Kalman filtering: Kalman filtering is performed on inertial navigation system and GPS data to eliminate random noise and improve the accuracy of position and velocity.

[0109] Median filtering: Perform median filtering on ultrasonic and lidar data to remove outliers (such as incorrect ranging caused by reflection interference).

[0110] 2.42 Anomaly Detection

[0111] Rule detection: Set a reasonable threshold range (such as wind speed, distance), and mark data outside the range as abnormal.

[0112] Machine learning detection: Use historical data to train anomaly detection models (such as support vector machines (SVMs) or isolation forest algorithms) to identify complex abnormal patterns.

[0113] 2.43 Data Standardization

[0114] Normalization processing: Normalize the data from different sensors to the same numerical range to facilitate subsequent fusion and analysis.

[0115] Time interpolation: Interpolate data with different sampling frequencies to ensure consistency in the time dimension.

[0116] 3. Self-learning control module

[0117] like Figure 3 As shown, it is used to select a deep neural network and a long short-term memory network based on a multi-task learning framework, and then use the objective function to train the model, generate a control strategy generation model, and update the control strategy generation model according to real-time data. The self-learning control module includes:

[0118] 3.1 Data Collection Unit

[0119] It is used to collect the multimodal data, environmental data (including wind speed, wind direction, wave height, tide direction, meteorological conditions, etc.) and ship operation data (manual operation records of crew members, such as emergency collision avoidance, speed adjustment, etc.), and perform data cleaning, normalization and feature extraction on the collected data to obtain a comprehensive data set.

[0120] Cleaning: Remove outliers and noisy data (such as erroneous data caused by sensor failure).

[0121] Normalization: Normalize data of different dimensions (such as speed and fuel consumption) to a unified range to facilitate model training.

[0122] Feature extraction: Extract key variables (such as fuel efficiency, collision avoidance distance, heading deviation, etc.) through feature engineering to reduce data dimensions.

[0123] 3.2 Model building unit

[0124] It is used to obtain an initial model based on a multi-task learning framework (training a model to simultaneously optimize multiple objectives, such as fuel efficiency and safety, and improving the generalization ability of the model), using a deep neural network to process nonlinear data in the comprehensive data set (such as the relationship between fuel consumption and speed, and thruster power), and using a long short-term memory network to process time series data in the comprehensive data set (such as the impact of environmental changes on control strategies during navigation).

[0125] 3.3 Model Training Unit

[0126] It is used to combine the objective function, use the multi-source sensor data, and train the initial model in a supervised learning manner to obtain a control strategy generation model.

[0127] 3.4 Model Update Unit

[0128] It is used to perform incremental learning of the control strategy generation model using real-time data to achieve parameter optimization.

[0129] 4. Ship digital model module

[0130] like Figure 3As shown, it is used to construct a physical model and an environmental model of a ship, obtain a digital model simulating the navigation of the ship, and use the digital model to test different ship control strategies. The ship digital model module includes:

[0131] 4.1 Digital model building unit

[0132] It is used to construct a physical model of the ship (including hull structure, propulsion system, steering gear system, etc.), introduce fluid mechanics into the physical model, simulate the dynamic characteristics of the ship under different speeds and sea conditions, construct an external environment model (such as wind, waves, tides, and the dynamic behavior of other ships), and generate a simulated navigation scene.

[0133] 4.2 Strategy Optimization Unit

[0134] It is used to combine the ship physical model and the simulated navigation model to perform scenario simulation to test the effectiveness of different ship control strategies, and optimize the control strategy generation model according to the test results.

[0135] 5. Decision execution module

[0136] The decision-making execution module is used to convert the output of the control strategy generation model into a control instruction and send the control instruction to each actuator of the ship.

[0137] 5.1 Instruction Generation Module

[0138] It is used to convert the output of the control strategy generation model into a control instruction with a unified format (in accordance with a standardized communication protocol (such as CAN bus, Modbus, to ensure compatibility with the actuator), and send the control instruction to the corresponding actuator through a communication network (such as industrial Ethernet or wireless communication) and a redundant communication mechanism; the control instruction includes a rudder angle adjustment instruction, a thruster power distribution instruction, a speed adjustment instruction and an auxiliary equipment control instruction. For example:

[0139] Steering gear control: According to the rudder angle adjustment command, the steering gear actuator accurately adjusts the rudder angle to ensure that the ship sails according to the predetermined course. The rudder angle feedback control mechanism is introduced to correct the execution error by real-time monitoring of the rudder angle position.

[0140] Thruster control: According to the thruster power allocation instructions, the speed and direction of the thrusters are adjusted to achieve precise control of the ship's speed and attitude. In dynamic positioning mode, the thrusters can be independently controlled to keep the ship in a fixed position or on a specific trajectory.

[0141] Auxiliary equipment control: Control auxiliary equipment such as ballast water system and fin stabilizers to optimize the stability and energy efficiency of the ship.

[0142] 5.2 Execution Feedback Module

[0143] It is used to adjust the output of the control strategy generation model by feeding back the execution result of the control instruction to the control strategy generation model.

[0144] 6. Safety monitoring module

[0145] It is used to detect abnormal behavior of ships and make graded alarms and records based on the detection results. The safety monitoring module includes:

[0146] 6.1 Anomaly Detection Unit

[0147] It is used to monitor ship operating parameters (speed, heading, fuel consumption, rudder angle, propeller power, etc.), environmental parameters (wind speed, wave height, obstacle distance, etc.) and equipment status (main engine temperature, pressure, vibration, etc.) in real time using a variety of sensors and edge computing devices, and perform anomaly detection according to preset abnormal ranges (such as normal ranges of speed and fuel consumption, which are marked as abnormal when exceeding the range) to obtain abnormal detection results, and then classify the abnormal detection results into low-risk anomalies, medium-risk anomalies and high-risk anomalies according to the severity of the abnormal detection results.

[0148] Low-risk anomalies: If fuel efficiency decreases, optimization suggestions will be provided.

[0149] Medium-risk exception: If the device performance deteriorates, a prompt is issued to check the device.

[0150] High-risk anomalies: such as approaching obstacles and equipment failures, trigger alarms.

[0151] 6.2 Alarm unit

[0152] It is used to preset alarm trigger conditions and perform graded alarms according to the classification of the abnormal detection results.

[0153] The alarm triggering conditions include: the distance between the ship and the obstacle is lower than the safety threshold, the main engine temperature or pressure exceeds the safety range, and the ship deviates from the scheduled route.

[0154] The graded alarms include:

[0155] Level 1 alarm, for the above-mentioned high-risk anomalies, such as approaching obstacles and equipment failures, triggers audible and visual alarms and recommends that the crew take over control.

[0156] Level 2 alarm, for the above-mentioned medium-risk anomalies, such as low fuel efficiency, the interface prompts optimization suggestions.

[0157] Level 3 alarms, for low-risk anomalies such as equipment performance degradation, prompt regular inspections.

[0158] The system will record all alarm events, including trigger time, risk level, response measures, etc.

[0159] Therefore, the present invention provides a high-efficiency ship automatic control system, which can not only integrate multiple subsystem modules to achieve data sharing, collaborative optimization and global control, but also provide optimal control strategies, reduce the cost and risk of actual navigation, and improve the ship's control accuracy, response speed and operating efficiency.

[0160] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0161] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A high-efficiency ship automatic control system, characterized in that: It includes an integrated control platform and an intelligent sensor network module, a self-learning control module, a ship digital model module, a decision execution module and a safety monitoring module connected to the integrated control platform; The integrated control platform is used to integrate the heading control subsystem, speed control subsystem, dynamic positioning subsystem, collision avoidance subsystem and energy efficiency management subsystem, and set the objective function to carry out data sharing, collaborative optimization and global control among the subsystems; The intelligent sensor network module is used to use multiple sensors to perform data collection and data fusion to obtain raw data, perform data preprocessing on the raw data to obtain multimodal data, and use edge computing devices to perform real-time data collection; The self-learning control module is used to select a deep neural network and a long short-term memory network based on a multi-task learning framework, and then use the objective function to perform model training, generate a control strategy generation model, and update the control strategy generation model according to real-time data; The ship digital model module is used to construct a physical model and an environmental model of the ship, obtain a digital model simulating the navigation of the ship, and use the digital model to test different ship control strategies; The decision execution module is used to convert the output of the control strategy generation model into a control instruction, and send the control instruction to each execution mechanism of the ship; The safety monitoring module is used to detect abnormal behavior of the ship and to perform graded alarms and records according to the detection results.

2. A high-efficiency ship automatic control system according to claim 1, characterized in that: The integrated control platform comprises: Data linkage unit, which is used to design data bus and communication protocol for data sharing between subsystems, standardize the data of each subsystem, synchronize subsystem data through timestamp mechanism, and achieve data consistency; A subsystem linkage unit is used to adjust the speed output power through the heading control target to realize the linkage between the heading control subsystem and the speed control subsystem, to realize the linkage between the dynamic positioning subsystem and the collision avoidance subsystem by introducing collision avoidance logic into the dynamic positioning subsystem, to adjust the speed and propeller power by monitoring the fuel consumption and emission data, and to realize the global optimization of the energy efficiency management subsystem; The collaborative optimization unit is used to define the objective function with fuel consumption, safety cost and navigation efficiency as the optimization targets and perform global optimization control; The human-computer interaction unit is used to design the user interface, display the ship's operating status and environmental information in real time, and present the data intuitively through dashboards, heat maps or navigation path maps.

3. A high-efficiency ship automatic control system according to claim 2, characterized in that: The definition process of the objective function includes: setting the minimum fuel consumption and the balance point between fuel efficiency and navigation time as optimization targets, and setting the collision avoidance distance and heading deviation as safety constraints, combining the optimization target and the safety constraints to obtain the objective function, and introducing a genetic algorithm or a particle swarm optimization algorithm into the objective function to calculate the multi-objective optimization problem; the formula of the objective function is: F=ω1·C f +ω2·S c +ω3·E t Among them, F is the objective function value, C f is the fuel consumption, S c is the security cost function, E t is the navigation efficiency cost function, ω1, ω2, ω3 are all weight coefficients, ω1+ω2+ω3=1.

4. A high-performance ship automatic control system according to claim 3, characterized in that: The calculation formula of the fuel consumption is: Wherein, P(t) is the output power of the ship's main engine at time t, SFC is the specific fuel consumption rate of the main engine, and T is the actual sailing time; The calculation formula of the security cost function is: Where N is the total number of obstacles during navigation, d i is the distance between the ship and the i-th obstacle, Δθ i is the deviation between the current heading of the ship and the planned heading for collision avoidance, k i is the weight coefficient used to balance the importance of collision avoidance distance and heading deviation; The calculation formula of the navigation efficiency cost function is: Where T is the actual voyage time, T optimal is the ideal sailing time, C f is the actual fuel consumption, C f·optimal is the ideal fuel consumption, α is the time cost weight coefficient used to represent the importance of time efficiency, and β is the fuel cost weight coefficient used to represent the importance of fuel efficiency.

5. A high-efficiency ship automatic control system according to claim 4, characterized in that: The intelligent sensor network module comprises: Historical data acquisition unit, used to deploy lidar, ultrasonic sensors, inertial navigation systems, meteorological sensors, GPS, radar and AIS to perform environmental perception and ship status monitoring and obtain multi-source sensor data; A data fusion unit is used to synchronize the multi-source sensor data with the timestamp, and to align the multi-source sensor data with the unified coordinate system, and then combine the distance data of the laser radar and ultrasonic sensor to generate obstacle information to complete low-level fusion, and combine the GPS, radar and AIS data to generate a ship distribution map to complete high-level fusion to obtain the original data; A real-time data collection unit is used to deploy multiple edge computing devices in different areas of the ship in a combination of wired and wireless methods to collect real-time data; The data preprocessing unit is used to perform noise filtering, anomaly detection and data standardization on the raw data and real-time data to obtain multimodal data.

6. A high-efficiency ship automatic control system according to claim 5, characterized in that: The self-learning control module comprises: A data collection unit, used to collect the multimodal data, environmental data and ship operation data, and perform data cleaning, normalization and feature extraction on the collected data to obtain a comprehensive data set; A model building unit, used to process the nonlinear data in the comprehensive data set using a deep neural network and process the time series data in the comprehensive data set using a long short-term memory network based on a multi-task learning framework to obtain an initial model; A model training unit, used to train the initial model in a supervised learning manner using the multi-source sensor data in combination with the objective function to obtain a control strategy generation model; The model updating unit is used to perform incremental learning of the control strategy generation model using real-time data to achieve parameter optimization.

7. A high-efficiency ship automatic control system according to claim 6, characterized in that: The ship digital model module comprises: A digital model building unit is used to build a physical model of the ship, introduce fluid mechanics into the physical model, simulate the dynamic characteristics of the ship under different speeds and sea conditions, build an external environment model, and generate a simulated navigation scene; A strategy optimization unit is used to perform scenario simulation in combination with the ship physical model and the simulated navigation model to test different ship control strategies, and optimize the control strategy generation model according to the test results.

8. A high-efficiency ship automatic control system according to claim 7, characterized in that: The decision execution module includes: An instruction generation module is used to convert the output of the control strategy generation model into a control instruction with a unified format, and send the control instruction to the corresponding actuator through a communication network and a redundant communication mechanism; the control instruction includes a rudder angle adjustment instruction, a thruster power distribution instruction, a speed adjustment instruction and an auxiliary equipment control instruction; The execution feedback module is used to adjust the output of the control strategy generation model by feeding back the execution result of the control instruction to the control strategy generation model.

9. A high-efficiency ship automatic control system according to claim 8, characterized in that: The safety monitoring module comprises: An anomaly detection unit is used to monitor ship operation parameters, environmental parameters and equipment status in real time using a variety of sensors and edge computing devices, and perform anomaly detection according to a preset anomaly range to obtain an anomaly detection result, and then classify the anomaly detection result into low-risk anomaly, medium-risk anomaly and high-risk anomaly according to the severity of the anomaly detection result; An alarm unit is used to preset alarm trigger conditions and perform graded alarms according to the classification of the abnormality detection results; the alarm trigger conditions include the distance between the ship and the obstacle being lower than a safety threshold, the main engine temperature or pressure exceeding a safety range, and the ship deviating from the scheduled route; the graded alarms include a level one alarm for the high-risk abnormalities, a level two alarm for the medium-risk abnormalities, and a level three alarm for the low-risk abnormalities.

Images