A ship traffic flow coordination method and system based on model predictive control

Abstract

The present invention discloses a ship traffic flow coordination method and system based on model predictive control, the method comprising the following steps: establishing a state prediction model, obtaining the real-time navigation state of the own ship and the oncoming ship, predicting the navigation state of the own ship and the oncoming ship in multiple future time steps according to the real-time navigation state; calculating the meeting point and meeting time of the own ship and the oncoming ship according to the predicted navigation state, establishing an optimization function with the goal of minimizing the ship's navigation time and fuel consumption; setting the constraint conditions of the optimization function so that the meeting point is located outside the dangerous waters and the meeting time is located outside the dangerous time window, the dangerous time window being the time period when the oncoming ship travels in the dangerous waters; solving the optimization function to obtain the optimal control parameters of the own ship, and updating the navigation state of the own ship in the next time step according to the optimal control parameters; repeating the above steps until the own ship arrives at the destination.

Description

Technical Field

[0001] The present invention relates to the field of ship traffic management and control, and in particular to a ship traffic flow coordination method and system based on model predictive control. Background Art

[0002] With the rapid development of the global shipping industry, the density of water traffic has increased significantly, and ships are frequently involved in accidents, which poses great challenges to navigation safety and traffic efficiency. Traditional ship traffic management methods mainly rely on manual command and navigation strategies based on fixed speeds, which are difficult to cope with complex and dynamic water traffic environments.

[0003] At present, some advanced ship navigation systems have begun to use automated technologies to assist navigation decisions. These systems usually include automatic collision avoidance systems (ACAS), vessel traffic management systems (VTMS), and rule-based navigation optimization algorithms. These technologies can predict potential collision risks to a certain extent and provide collision avoidance suggestions.

[0004] Although existing technologies have made some progress in ship navigation safety, there are still some limitations. These systems often lack sufficient flexibility and adaptability when dealing with complex and changing marine environments and traffic conditions, and it is difficult to achieve real-time dynamic adjustments, resulting in the safety of the optimization results not being guaranteed. Therefore, a more advanced and comprehensive solution is needed to overcome these problems and improve the safety and economy of ship navigation. Summary of the invention

[0005] The present invention proposes a ship traffic flow coordination method and system based on model predictive control, which solves the problem of poor flexibility and adaptability of existing ship traffic flow planning methods.

[0006] In order to solve the above technical problems, the present invention provides a ship traffic flow coordination method based on model predictive control, comprising the following steps:

[0007] Step S1: Establish a state prediction model, obtain the real-time navigation state of the own ship and the oncoming ship, and predict the navigation state of the own ship and the oncoming ship in multiple future time steps according to the real-time navigation state;

[0008] Step S2: Calculate the meeting point and time of the own ship and the oncoming ship according to the predicted navigation state, and establish an optimization function with the goal of minimizing the navigation time and fuel consumption of the own ship;

[0009] Step S3: setting the constraint conditions of the optimization function so that the encounter point is outside the dangerous waters and the encounter time is outside the dangerous time window, where the dangerous time window is the time period when the oncoming ship is traveling in the dangerous waters;

[0010] Step S4: solving the optimization function to obtain the optimal control parameters of the ship, and updating the navigation state of the ship at the next time step according to the optimal control parameters;

[0011] Step S5: Repeat steps S1 to S4 until the ship reaches the destination.

[0012] Preferably, the expression of the state prediction model in step S1 is:

[0013] ;

[0014] ;

[0015] ;

[0016] ;

[0017] In the formula, for t Moment of the future The navigation status of the ship within a time step; is the state transfer matrix k Power; Indicates the position of the vessel; is the control input matrix; for Control parameters of the ship at each moment; for t The ship's navigation status at the moment; Indicates the speed of the ship; represents the acceleration of the ship; is a time step.

[0018] Preferably, the expression for calculating the encounter time in step S2 is:

[0019] ;

[0020] In the formula, The meeting time between own ship and the oncoming ship; , are the speeds of own ship and oncoming ship respectively; The distance between own ship and the oncoming ship.

[0021] Preferably, in step S2, the position of the meeting point is determined by calculating the distance between the meeting point and the starting point of the own ship, and the expression for calculating the distance between the meeting point and the starting point of the own ship is:

[0022] ;

[0023] In the formula, The distance between the meeting point of own ship and the oncoming ship and the starting point of own ship The meeting time between own ship and the oncoming ship; The ship's speed.

[0024] Preferably, the expression of the optimization function in step S2 is:

[0025] ;

[0026] ;

[0027] ;

[0028] In the above formula, F is the optimization function; is the change in own ship’s speed; The sailing time of the vessel; is the fuel consumption of own ship; is the acceleration penalty term; Penalty item for the meeting point; , , , , is the weight coefficient; is a constant; is the ship's speed; is the adjustment factor for fuel consumption; , is the weight coefficient; is the acceleration of the ship; For the current moment.

[0029] Preferably, the constraint condition in step S3 is expressed as:

[0030] ;

[0031] ;

[0032] ;

[0033] In the above formula, The position of the meeting point between own ship and the oncoming ship; The distance between the dangerous waters and the starting point of the ship; , The distances from the starting point and the end point of the dangerous waters to the starting point of the ship respectively; The meeting time between own ship and the oncoming ship; , are the start and end times of the dangerous time window respectively.

[0034] Preferably, solving the optimization function in step S4 includes the following steps:

[0035] Step S41: Select a point outside the dangerous waters as a new meeting point, and calculate a new meeting time based on the preset meeting point:

[0036] ;

[0037] In the formula, The preset meeting point The corresponding meeting time; is the preset meeting point, , is a minimum value, is the distance between the starting point of the dangerous waters and the starting point of the ship. is the length of the current water area; is the ship’s speed;

[0038] Step S42: Calculate the acceleration of the ship at the current encounter point:

[0039] ;

[0040] ;

[0041] ;

[0042] In the above formula, The ship speed of the current encounter point; is the distance between own ship and the oncoming ship; is the speed increment of own ship; is the speed of the oncoming ship; is the acceleration of own ship;

[0043] Step S43: Calculate the value of the optimization function of the ship under the current acceleration;

[0044] Step S44: Repeat steps S41 to S43, and select the acceleration with the smallest value of the optimization function as the optimal control parameter of the ship.

[0045] The present invention also provides a ship traffic flow coordination system based on model predictive control, which is implemented based on the above-mentioned ship traffic flow coordination method based on model predictive control, and includes a data acquisition module, a state prediction module, a parameter optimization module and a traffic coordination module;

[0046] The data acquisition module is used to collect status parameters of all ships in the water area;

[0047] The state prediction module predicts the navigation state of all ships in the water area in multiple future time steps according to the state parameters collected by the data collection module;

[0048] The parameter optimization module calculates the control parameters of all ships in the next time step according to the navigation state predicted by the state prediction module, so that the meeting point of every two ships in the water area is located outside the dangerous water area;

[0049] The traffic coordination module adjusts the status of all ships in the next time step according to the parameters output by the parameter optimization module.

[0050] The present invention also provides an electronic device, comprising: a memory, a processor and a computer program, wherein the computer program is stored in the memory and is configured to be executed by the processor to implement the above-mentioned ship traffic flow coordination method based on model predictive control.

[0051] The present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned ship traffic flow coordination method based on model predictive control.

[0052] The benefits of the present invention include at least:

[0053] 1. By predicting the future navigation status of own ship and oncoming ship, potential collision risk can be identified in advance. By calculating the meeting point and meeting time and ensuring that they are outside the dangerous waters and dangerous time window, it is possible to avoid encountering oncoming ships under dangerous conditions, thereby improving navigation safety;

[0054] 2. Establishing an optimization function with the goal of minimizing the ship's sailing time and fuel consumption helps improve navigation efficiency and reduce unnecessary fuel consumption. By acquiring and predicting the status in real time, navigation decisions can be dynamically adjusted according to the latest navigation status, improving the ability to respond to emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of a method flow of an embodiment of the present invention;

[0056] Figure 2 A schematic diagram of a simulation process of an embodiment of the present invention;

[0057] Figure 3 A schematic diagram of dual-ship path planning according to an embodiment of the present invention;

[0058] Figure 4 Schematic diagram of multi-vessel path planning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings 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 belong to the protection scope of the present invention.

[0060] Model Predictive Control (MPC) is an advanced control strategy. Its basic principle is to use the dynamic model of the system to predict the future system behavior and achieve the optimization goal by optimizing the current control input. The Green Wave Theory was originally used for traffic control. Its main purpose is to optimize road traffic flow and reduce congestion and parking time. The theory controls the timing of traffic signals so that vehicles can connect green lights between a series of signal lights when traveling at a specific speed, thereby achieving barrier-free passage.

[0061] In a complex scenario with multiple ships and multiple dangerous waters, ensuring that ships meet in non-dangerous waters is a high-dimensional, nonlinear problem. To this end, this patent proposes a new optimization control strategy - Model Green Wave Predictive Control (MGPC). This method combines model predictive control with green wave theory and is a new model-based optimization control strategy.

[0062] In ship traffic, although there are no traffic lights, there are dangerous waters where the accident rate is significantly higher than that in ordinary waters, such as narrow waterways, multi-directional traffic intersections, and waters with fixed obstacles. By adjusting the speed of the ship, two ships can meet safely in non-dangerous waters or within the green wave time window, which can avoid collisions or slow down waiting in dangerous areas.

[0063] Based on the above theory, the embodiment of the present invention provides a Figure 1 The ship traffic flow coordination method based on model predictive control shown in the figure, taking the case of two ships as an example, includes the following steps:

[0064] Step S1: Establish a state prediction model, obtain the real-time navigation status of the ship and the oncoming ship, and predict the navigation status of the ship and the oncoming ship in multiple future time steps based on the real-time navigation status.

[0065] Specifically, a model that can describe the dynamic behavior of the ship is first established. This model includes state variables such as the ship's speed, position, acceleration, etc. The following state equation is used in the embodiment of the present invention to describe the motion state of the ship:

[0066] ;

[0067] in, Indicates ship location; Indicates ship speed; Indicates ship acceleration.

[0068] The state equation can also be expressed as:

[0069] ;

[0070] in, Indicates ship The control input is used to adjust the acceleration of the ship.

[0071] In order to ensure the coordinated navigation of multiple ships, the relative positions and speeds of the ships must be monitored and optimized at all times. and ships The relative position between and relative speed It can be defined as:

[0072] ;

[0073] ;

[0074] These relative positions and speed relationships will be used to construct constraints for multi-ship collaboration to ensure that ships maintain a safe distance.

[0075] In the MGPC algorithm, the system is usually continuous time, but the system needs to be discretized during the implementation process. The discretization process is to convert the state space model of the continuous time system into a discrete time form to facilitate the implementation of the algorithm on the computer. MGPC predicts the future state of the ship through the current state of the ship, such as the speed and position of each ship, and the control input of the ship at the future moment, such as the speed adjustment. Based on the predicted ship state, MGPC can determine whether the ship will enter dangerous waters or collide with other ships in the future.

[0076] In discrete time, the system can be discretized as:

[0077] ;

[0078] ;

[0079] ;

[0080] in, Ais the state transfer matrix; B is the control input matrix; is a time step.

[0081] Through this discretization form, the motion state of the ship in multiple time steps in the future can be predicted, so as to determine whether each ship will enter dangerous waters or collide with other ships in multiple time steps in the future. , predicting the future The core of the prediction model relies on the state equation:

[0082] ;

[0083] In the formula, for t Moment of the future The navigation status of the ship within a time step; is the state transfer matrix k Power, used to describe the system k How the state after time steps evolves based on the current state; is the control input matrix, which is used to convert the control input Mapping to state changes reflects the impact of control input on the system state.

[0084] Step S2: Calculate the meeting point and time of the own ship and the oncoming ship according to the predicted navigation status, and establish an optimization function with the goal of minimizing the navigation time and fuel consumption of the own ship.

[0085] Specifically, the expression for calculating the encounter time between own ship and oncoming ship is:

[0086] ;

[0087] In the formula, The meeting time between own ship and the oncoming ship; , are the speeds of own ship and oncoming ship respectively; is the distance between own ship and oncoming ship.

[0088] The position of the encounter point is determined by calculating the distance between the encounter point and the starting point of the ship. The expression for calculating the distance between the encounter point and the starting point of the ship is:

[0089] ;

[0090] In the formula, It is the distance between the meeting point of own ship and the oncoming ship and the starting point of own ship.

[0091] The core goal of MGPC optimization is to adjust the speed of the ship so that the ship can effectively avoid encountering other ships in dangerous waters. On this basis, multiple optimization goals are further pursued: first, to ensure that the fuel consumption of the ship is reduced as much as possible during navigation, thereby improving fuel efficiency; second, on the premise of ensuring safe navigation, to shorten the overall navigation time as much as possible and improve navigation efficiency; finally, to smoothly control the adjustment of the ship's speed to limit excessive acceleration and deceleration, thereby reducing the mechanical wear and energy consumption of the ship, and ensuring the economy and environmental protection of navigation. Therefore, this patent comprehensively considers multiple factors such as encounter time, acceleration, fuel consumption and navigation time, and constructs an optimization function that balances safety, economy and environmental protection. This optimization function not only avoids the encounter of ships in dangerous areas, but also needs to achieve trade-offs and coordination between multiple optimization goals, so as to improve the overall navigation efficiency and safety.

[0092] The expression of the optimization function finally constructed is:

[0093] ;

[0094] ;

[0095] ;

[0096] In the above formula, F is the optimization function; is the change in own ship’s speed; The sailing time of the vessel; is the fuel consumption of own ship; is the acceleration penalty term; Penalty item for the meeting point; , , , , is the weight coefficient; is a constant; is the ship's speed; is the adjustment factor for fuel consumption; , is the weight coefficient; for the current moment;

[0097] The greater the acceleration and the more frequent the speed changes, the greater the acceleration penalty term. At the same time, the penalty term of the encounter point is set to be two orders of magnitude higher than the values ​​of other penalty functions to ensure that the encounter point is not in dangerous waters.

[0098] Step S3: setting the constraints of the optimization function so that the meeting point is outside the dangerous waters and the meeting time is outside the dangerous time window. The dangerous time window is the time period when the oncoming ship is traveling in the dangerous waters.

[0099] In the field of road traffic control, the application of green wave theory not only relies on the coordination of traffic signal light sequences, but also needs to comprehensively consider the vehicle speed adjustment to ensure that vehicles can continuously pass through green light sections. By optimizing the matching of signal cycle and vehicle speed, traffic delays can be effectively reduced, traffic efficiency can be improved, and a more ideal traffic flow state can be achieved.

[0100] However, in water traffic, ships lack signal control and are affected by complex environments such as channel width, flow velocity and wind direction. Therefore, it is difficult to apply the green wave theory to water traffic. Especially in dangerous waters, the speed and encounter risk of different ships are difficult to coordinate and control. Therefore, there are many challenges in achieving coordinated passage similar to green waves in a water environment without signal lights. In water traffic, due to the combined effects of multiple factors such as channel width, curvature, water flow conditions and wind force, algorithms can be used to designate certain waters as dangerous waters. In these areas, if ships encounter encounters, overtaking or crossing encounters, it will have a significant impact on navigation safety. In addition, when ships meet in dangerous waters, they usually take deceleration measures to prevent navigation accidents, which not only increases the difficulty of ship drivers' control, but also reduces traffic efficiency.

[0101] Therefore, it is particularly important to avoid encounters between ships in dangerous waters. In the field of navigation management and optimization, the application of the green wave theory has been innovatively expanded, aiming to achieve efficient and safe passage in non-restricted waters by finely controlling the speed of ships. Specifically, the green wave time window for ship navigation is constructed by analogy with the green light time interval in road traffic. This window defines a time interval within which ships can safely cross dangerous waters without encountering other ships. The core of this method is to accurately calculate and adjust the navigation speed of the ship through advanced algorithms and prediction models to ensure that it can accurately match and enter the green wave time window in a timely manner, thereby effectively avoiding the risk of encountering other ships in potentially dangerous waters.

[0102] Therefore, in the embodiment of the present invention, the constraint condition is set as:

[0103] ;

[0104] In the formula, The position of the meeting point between own ship and the oncoming ship; The distance between the dangerous waters and the starting point of the ship; , They are the distances from the starting point and the end point of the dangerous waters to the starting point of the ship.

[0105] Applying the idea of ​​green wave time window, the constraint condition of the encounter time is set as:

[0106] ;

[0107] ;

[0108] In the formula, The meeting time between own ship and the oncoming ship; , are the start and end times of the dangerous time window respectively.

[0109] Will The value is set to:

[0110] ;

[0111] In the formula, The penalty value of the judgment function for determining whether two ships will meet in dangerous waters should be at least two orders of magnitude higher than the values ​​of other items in the optimization function to reflect its decisive position.

[0112] Step S4: Solve the optimization function to obtain the optimal control parameters of the ship, and update the navigation state of the ship in the next time step according to the optimal control parameters. Solving the optimization function includes the following steps:

[0113] Step S41: Select a point outside the dangerous waters as a new meeting point, and calculate a new meeting time based on the preset meeting point:

[0114] ;

[0115] In the formula, is the preset meeting point, , is a minimum value, is the distance between the starting point of the dangerous waters and the starting point of the ship. is the length of the current water area; The preset meeting point The corresponding meeting time; The speed of this ship.

[0116] Step S42: Calculate the acceleration of the ship at the current encounter point:

[0117] ;

[0118] ;

[0119] ;

[0120] In the above formula, The ship speed of the current encounter point; is the distance between own ship and the oncoming ship; is the speed increment of own ship; is the speed of the oncoming ship; is the acceleration of own ship.

[0121] Step S43: Calculate the value of the optimization function of the ship under the current acceleration.

[0122] Step S44: Repeat steps S41 to S43, and select the acceleration with the smallest value of the optimization function as the optimal control parameter of the ship.

[0123] Step S5: Repeat steps S1 to S4 until the ship reaches the destination.

[0124] Specifically, at each time step, MGPC will predict the system state in the next few time steps based on the current state, solve the ship's state parameters according to the optimization function, give the optimal control strategy, and then apply the first control parameter to control the ship's speed in the next time step, performing rolling optimization and real-time adjustment of the ship's control parameters.

[0125] In order to verify the effectiveness of the method of the present invention, simulation and experiments are performed to verify its effect. Figure 2 As shown, the following steps are included:

[0126] 1. Obtain AIS data of ships in the research waters. AIS data includes the ship's MMSI code, time, longitude and latitude, speed, heading and other information. By preprocessing the AIS data in the research waters, including data matching, track segmentation, data cleaning and missing value completion, we can ensure that the data accurately reflects the dynamics of the ship and provide support for ship speed adjustment and related research. AIS data includes MMSI code, time, longitude and latitude, speed, heading and other information. ,longitude ,latitude ,speed ,course The navigation information of a ship at any time can be abbreviated as:

[0127] .

[0128] 2. Calculate the daily trend of the number of ships through time series analysis. Use the mean and variance to describe the change in the number of ships, providing basic data support for speed control and traffic optimization.

[0129] ;

[0130] ;

[0131] in, represents the average number of ships; represents the variance of the number of ships; Indicates i Number of ships per day.

[0132] By analyzing historical AIS data, we can find out the changing patterns of the number of ships in different time periods, which provides basic data support for the establishment of the simulation model.

[0133] 3. Use the probability density function (PDF) to analyze the law of ships arriving at the study waters in a specific time period. PDF can be used to obtain the probability distribution of ships entering the study waters at different times, which is of great significance for understanding the traffic flow in the study waters and establishing a simulation model.

[0134] The mathematical definition of PDF is: If the distribution function of a random variable X in the interval [a, b] is , then in x The probability density function at Defined as The derivative of :

[0135] ;

[0136] After determining the research waters, the longitude and latitude data in the ship AIS data are used as the screening basis to screen out all the arrival time information of ships entering the research waters, and the probability density function is used to analyze the arrival time distribution of the ships. The kernel density estimation method is used to fit the arrival time to find the peak arrival time of the ship in a day:

[0137] ;

[0138] in, is the sample data In bandwidth The kernel density estimate under ; is the kernel function; is the sample size; is the sample data of arrival time; is the target estimation point.

[0139] 4. Add ship domain to the ship traffic flow simulation model. In the ship traffic flow simulation model, the ship domain refers to the minimum safe distance between a ship and other ships during navigation. This distance can be determined based on the minimum interval when ships pass each other in different directions. By observing the minimum passing distance when ships meet in different directions during navigation, the scope of the ship's safety domain can be obtained. The ship domain in the simulation model is determined based on this statistical data result to ensure the safe distance between ships and avoid collisions. The model of this ship domain is calculated by counting the relative positions and passing distances of different ships in the navigation in AIS data. Based on historical AIS data, by counting the minimum passing distance when ships meet in different directions during navigation, and then performing data density analysis on the statistical results, the distance value with the largest data density is selected as the safe distance in that direction, and the ship domain is established based on this. In this way, a suitable domain size can be set for the ship during the simulation process, so that in the simulation environment, the ship can maintain a safe distance when approaching other ships.

[0140] 5. Build a simulation model. In the study of ship traffic flow optimization control, it is crucial to build a simulation model that can truly reflect the dynamic behavior of actual traffic flow. Through this simulation model, the dynamic behavior of different ships in the waters can be simulated, providing accurate theoretical support and data basis for subsequent speed optimization, danger avoidance, and ship green wave control.

[0141] By processing and counting AIS data, we can find out the changing patterns of ship traffic in different time periods and different waters, and then provide input parameters for the simulation model to accurately simulate the actual navigation situation. The simulation logic is as follows:

[0142] (1) Ship departure: The departure time and initial speed of the ship are randomly generated according to the statistical law of traffic flow. The departure time is simulated by uniform distribution or normal distribution, and the initial speed is generated according to the departure speed distribution of the ship.

[0143] (2) Acceleration logic: When in non-dangerous waters and when there are no other ships ahead, the ship will accelerate appropriately based on the difference between the speed limit and the current speed.

[0144] The acceleration can be calculated by the following formula

[0145] ;

[0146] in, The maximum safe speed of the ship is obtained through the statistical results of historical AIS data. Its value is the maximum speed of the ship when sailing in the waters; is the current speed of the ship; The time required to accelerate the vessel; is the maximum acceleration, and its value is set according to statistical results.

[0147] (3) Deceleration logic: In dangerous waters, the ship will slow down appropriately to ensure safety. Deceleration can be achieved by controlling the deceleration rate:

[0148] ;

[0149] in, is the minimum permissible speed in dangerous waters; is the deceleration time; is the maximum deceleration.

[0150] (4) Following logic: When a ship approaches the ship ahead and its current speed is greater than that of the ship ahead, the following ship should slow down appropriately and follow the ship ahead, as the collision avoidance rules prohibit overtaking in narrow waterways.

[0151] The following model can be borrowed from the form of Intelligent Driver Model (IDM), taking into account the distance and speed difference with the ship in front:

[0152] ;

[0153] in, is the current acceleration of the ship; It is the maximum acceleration that a ship can achieve under optimal conditions; is the desired speed, set according to the speed of the preceding ship; is the acceleration index; is the distance between the two ships; is the speed difference between own ship and preceding ship;

[0154] ) is the safety distance, defined as:

[0155] ;

[0156] in, is the minimum safe distance; For reaction time.

[0157] In a two-ship encounter scenario, the upstream ship is ship a, and the downstream ship is ship b. The two ships entered the controlled waters one after another, and both reported their navigation information to the VTS command center. After obtaining the information, it can be obtained through algorithm calculation that if the speed of the two ships is not adjusted, the two ships will meet in dangerous waters, which seriously threatens the navigation safety of the ships. The optimization goal of the present invention is to prevent the two ships from meeting in dangerous waters. By using the time period when the oncoming ship is traveling in the dangerous waters as the dangerous time window of the ship, and the time period when the oncoming ship is traveling outside the dangerous waters as the green wave time window of the ship, the speed of the two ships is optimized so that the two ships enter and exit the dangerous waters within their respective green wave time windows. After calculation, the optimized speed at which the two ships can avoid meeting in dangerous waters is obtained, and sent to the ship driver through the VTS center, thereby achieving the purpose of reducing navigation risks. The speed adjustment method for this scenario is as follows:

[0158] First, solve the meeting time of the two ships:

[0159] ;

[0160] In the formula, Indicates the time when the two ships will meet; It means finding the time that makes the equation valid; and Represent the speeds of ship a and ship b respectively.

[0161] The positions of two ships when they meet can be calculated based on the position of either ship. The distance from the starting point A is used to calculate the distance.

[0162] For upstream ship a:

[0163] ;

[0164] For downstream ship b:

[0165] ;

[0166] in, It indicates the distance from the starting point A when the two ships meet; Indicates the total length of the route in the study waters.

[0167] If this satisfies:

[0168] ;

[0169] in, Indicates the distance interval between the dangerous waters and the starting point A:

[0170] ;

[0171] In the formula, Indicates the distance between the starting point of the dangerous waters and the starting point A; Indicates the distance between the end point of the dangerous waters and the starting point A.

[0172] If the above conditions are met, the ship a that departs first will be allowed to maintain its speed. By adjusting the speed of the later departing ship b, the and , so that it satisfies .

[0173] Select a point outside the dangerous waters as the new meeting point, and solve the new meeting time according to the speed of the speed-maintaining ship a:

[0174] ;

[0175] in, Indicates safe waters. , Indicates a small safety distance.

[0176] Combine the new meeting time and the new meeting point to solve the velocity change and acceleration.

[0177] ;

[0178] ;

[0179] ;

[0180] In the above formula, Adjust the amplitude for the desired speed, a is the acceleration. The linear acceleration or deceleration formula can be used to calculate the specific adjustment plan. This formula ensures that the ship can avoid the dangerous area in time by adjusting the speed.

[0181] Calculate the value of the optimization function of ship b under the current acceleration, repeat the above steps, and select the acceleration with the smallest value of the optimization function as the optimal control parameter of ship b. The situations that ships a and b will encounter before and after adjustment are as follows: Figure 3 As shown, it can be seen that in the double-ship simulation, by using the method of the present invention to adjust the speed of the latter ship and obtain a new meeting time and meeting place, it can be ensured that the two ships will not meet in dangerous waters, so that the ships pass through dangerous waters within the green wave time window, avoiding accidents.

[0182] The method of the present invention is also applicable to the case of multiple ships. The situations encountered before and after the adjustment of multiple ships are as follows: Figure 4In the multi-ship simulation scenario, the MGPC algorithm adjusts the speed of each ship in real time through rolling optimization to ensure that the ships maintain a safe distance and avoid multiple ships entering dangerous waters at the same time. In the multi-ship scenario, the expression of the optimization function of the present invention is:

[0183] ;

[0184] in n is the total number of ships in the waters.

[0185] The expression of the acceleration penalty term is:

[0186] .

[0187] The method of the present invention optimizes the ship speed while taking into account the fuel consumption problem through the dynamic adjustment of the MPC algorithm. The simulation results show that through reasonable speed adjustment, unnecessary deceleration and acceleration processes can be reduced and fuel consumption can be reduced. At the same time, the concept of green wave time window is introduced, which enables ships to pass through dangerous waters at the optimal speed and avoid accidents in dangerous waters. The MGPC algorithm has real-time response capabilities and can dynamically optimize and control ships according to real-time AIS data to ensure the safety of ships in complex scenarios. The simulation results show that the ship traffic management system using this technology significantly improves the safety and traffic efficiency of water transportation when dealing with complex scenarios with multiple ships.

[0188] An embodiment of the present invention also provides a ship traffic flow coordination system based on model predictive control, which is implemented based on the above-mentioned ship traffic flow coordination method based on model predictive control, and includes: a data acquisition module, a state prediction module, a parameter optimization module and a traffic coordination module.

[0189] The data acquisition module is used to collect the status parameters of all ships in the water area.

[0190] The state prediction module predicts the navigation state of all ships in the water area in multiple future time steps based on the state parameters collected by the data acquisition module.

[0191] The parameter optimization module calculates the control parameters of all ships in the next time step according to the navigation state predicted by the state prediction module, so that the meeting point of every two ships in the water area is located outside the dangerous waters.

[0192] The traffic coordination module adjusts the status of all ships in the next time step according to the parameters output by the parameter optimization module.

[0193] An embodiment of the present invention also provides an electronic device, comprising: a memory, a processor and a computer program, wherein the computer program is stored in the memory and is configured to be executed by the processor to implement the above-mentioned ship traffic flow coordination method based on model predictive control.

[0194] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned ship traffic flow coordination method based on model predictive control.

[0195] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. Only the preferred embodiments of the present invention are expressed. The description is more specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. As long as there is no contradiction in the combination of these technical features, they should be considered as the scope recorded in this specification.

[0196] It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these modifications and improvements all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A ship traffic flow coordination method based on model predictive control, characterized in that: The following steps are involved: Step S1: Establish a state prediction model, obtain the real-time navigation state of the own ship and the oncoming ship, and predict the navigation state of the own ship and the oncoming ship in multiple future time steps according to the real-time navigation state; Step S2: Calculate the meeting point and time of the own ship and the oncoming ship according to the predicted navigation state, and establish an optimization function with the goal of minimizing the navigation time and fuel consumption of the own ship; The expression of the optimization function is: minF(Δv,T,fuel,C a ,D meet )=α1·Δv 2 +α2·T+α3·fuel+α4·C a +α5·D meet ; fuel=C·v 3 ·f; In the above formula, F is the optimization function; Δv is the speed change of the ship; T is the sailing time of the ship; fuel is the fuel consumption of the ship; C a is the acceleration penalty term; D meet is the penalty term for the encounter point; α1, α2, α3, α4, α5 are weight coefficients; C is a constant; v is the ship's speed; f is the adjustment factor for fuel consumption; β1, β2 are weight coefficients; a is the acceleration of the ship; t is the current moment; Step S3: setting the constraint conditions of the optimization function so that the encounter point is outside the dangerous waters and the encounter time is outside the dangerous time window, where the dangerous time window is the time period when the oncoming ship is traveling in the dangerous waters; Step S4: solving the optimization function to obtain the optimal control parameters of the ship, and updating the navigation state of the ship at the next time step according to the optimal control parameters; Step S5: Repeat steps S1 to S4 until the ship reaches the destination.

2. A ship traffic flow coordination method based on model predictive control according to claim 1, characterized in that: The expression of the state prediction model in step S1 is: Where X(t+k) is the navigation status of the ship in the next k time steps at time t; A k is the kth power of the state transfer matrix; x(t) represents the position of the ship; B is the control input matrix; u(t+j) is the control parameter of the ship at time t+j; X(t) is the navigation state of the ship at time t; v(t) represents the speed of the ship; a(t) represents the acceleration of the ship; Δt is a time step.

3. A ship traffic flow coordination method based on model predictive control according to claim 1, characterized in that: The expression for calculating the encounter time in step S2 is: In the formula, t meet is the meeting time of own ship and the oncoming ship; a (u), v b (u) are the speeds of own ship and oncoming ship respectively; d is the distance between own ship and oncoming ship.

4. A ship traffic flow coordination method based on model predictive control according to claim 1, characterized in that: In step S2, the position of the encounter point is determined by calculating the distance between the encounter point and the starting point of the own ship. The expression for calculating the distance between the encounter point and the starting point of the own ship is: In the formula, x meet is the distance between the meeting point of own ship and the oncoming ship and the starting point of own ship; t meet is the meeting time of own ship and the oncoming ship; a The ship's speed.

5. The ship traffic flow coordination method based on model predictive control according to claim 1, characterized in that: The expression of the constraint condition in step S3 is: t meet <t start ; t meet >t end ; In the above formula, x meet is the position of the meeting point between own ship and the oncoming ship; x danger is the distance between the dangerous waters and the starting point of the ship; x start 、x end are the distances from the starting point and the end point of the dangerous waters to the starting point of the ship; t meet is the meeting time of own ship and the oncoming ship; t start ,t end are the start and end times of the dangerous time window respectively.

6. A ship traffic flow coordination method based on model predictive control according to claim 1, characterized in that: Solving the optimization function in step S4 includes the following steps: Step S41: Select a point outside the dangerous waters as a new meeting point, and calculate a new meeting time based on the preset meeting point: Where, t i is the preset meeting point x i The corresponding meeting time; x i is the preset meeting point, x i ∈(0, x start -ε)∪(x start +ε, x), ε is a minimum value, x start is the distance between the starting point of the dangerous waters and the starting point of the ship, x is the length of the current waters; v is the speed of the ship; Step S42: Calculate the acceleration of the ship at the current encounter point: In the above formula, v' is the ship's speed at the current encounter point; d is the distance between the ship and the oncoming ship; Δv is the ship's speed increment; v other is the speed of the oncoming ship; a is the acceleration of the ship; Step S43: Calculate the value of the optimization function of the ship under the current acceleration; Step S44: Repeat steps S41 to S43, and select the acceleration with the smallest value of the optimization function as the optimal control parameter of the ship.

7. A ship traffic flow coordination system based on model predictive control, implemented based on a ship traffic flow coordination method based on model predictive control as claimed in any one of claims 1 to 6, characterized in that: include: Data acquisition module, state prediction module, parameter optimization module and traffic coordination module; The data acquisition module is used to collect status parameters of all ships in the water area; The state prediction module predicts the navigation state of all ships in the water area in multiple future time steps according to the state parameters collected by the data collection module; The parameter optimization module calculates the control parameters of all ships in the next time step according to the navigation state predicted by the state prediction module, so that the meeting point of every two ships in the water area is located outside the dangerous water area; The traffic coordination module adjusts the status of all ships in the next time step according to the parameters output by the parameter optimization module.

8. An electronic device comprising: A memory, a processor and a computer program, characterized in that: the computer program is stored in the memory and is configured to be executed by the processor to implement a ship traffic flow coordination method based on model predictive control as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement a ship traffic flow coordination method based on model predictive control as described in any one of claims 1 to 6.

Images