System and method for optimizing EEOI based on AI simulation navigation data model

Abstract

The invention discloses a system for optimizing EEOI based on an AI simulation navigation data model. The system comprises a data acquisition module, an AI model optimization analysis module, an AI autonomous operation module and a user active intervention module. The data acquisition module acquires fuel consumption data, energy efficiency parameters and external environment data of a ship in real time; the AI model optimization analysis module generates optimization suggestions for ship energy efficiency operation by processing real-time data information; the AI autonomous operation module receives an optimization suggestion of ship energy efficiency operation, actively intervenes in controlling the rotating speed of a ship main engine and the output of an on-board energy consumption device, then collects data again and continues to transmit the data to the AI model optimization analysis module for continuous optimization; and the user active intervention module is used for reminding manual intervention operation when the ship encounters an emergency. Through close cooperation of all the modules, a closed loop is formed in ship energy efficiency management from data acquisition and analysis to optimization suggestion, and the overall level of ship energy efficiency management is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of navigation control optimization, and in particular to a system and method for optimizing EEOI based on an AI-simulated navigation data model. Background Art

[0002] In the global shipping industry, ship energy efficiency management has become a key research topic and practical need. Traditionally, ships rely on single diesel engines for propulsion and energy efficiency management. While reliable, this approach has significant energy efficiency deficiencies, primarily manifested in the following areas:

[0003] High Fuel Consumption: The single diesel engine propulsion system fails to fully utilize modern energy efficiency optimization technologies, resulting in high fuel consumption. Speed control is usually determined by the crew based on experience, making it difficult to precisely maintain optimal fuel consumption. This significantly increases fuel costs, especially on long-distance voyages.

[0004] High pollution emissions: Diesel engines emit large amounts of pollutants during combustion, particularly greenhouse gases and hazardous substances such as carbon dioxide, nitrogen oxides, and sulfur compounds. With increasingly stringent environmental regulations and increasing restrictions on ship emissions, the high emissions of traditional ships are a growing concern.

[0005] Lack of effective real-time energy efficiency monitoring and feedback mechanisms: Traditional ship energy efficiency management systems mostly lack real-time monitoring and feedback functions, making it difficult to dynamically adjust operating parameters such as speed and route. As a result, energy efficiency optimization cannot be carried out in real time, thus missing the opportunity to save fuel through subtle adjustments. Summary of the Invention

[0006] Purpose of the invention: To solve the above problems, a system and method for optimizing EEOI based on an AI-simulated navigation data model are proposed. By integrating the optimization algorithm and the AI model, the optimized calculation of the ship's EEOI is realized; thereby, the energy efficiency optimization management and operation management of operating ships are realized, and the economic benefits of operating ships are improved.

[0007] Technical Solution: A system for optimizing EEOI based on an AI-simulated navigation data model, including a data acquisition module, an AI model optimization and analysis module, an AI autonomous operation module, and a user active intervention module. The data acquisition module is connected to ship sensors and monitoring equipment to collect real-time data on the ship's fuel consumption, energy efficiency parameters, and external environment.

[0008] The AI model optimization and analysis module calculates the parameters of ship energy efficiency operation by processing real-time data information, and generates optimization suggestions for ship energy efficiency operation based on the optimization analysis results;

[0009] After receiving the optimization suggestions for ship energy efficiency operation, the AI autonomous operation module actively intervenes to control the ship's main engine speed and the output of the ship's energy consumption devices. It then collects real-time navigation data based on the optimization suggestions and continues to pass it to the AI model optimization analysis module for further optimization. The optimized data is then submitted to the AI autonomous operation module.

[0010] The user active intervention module will prompt manual intervention when the ship encounters an emergency.

[0011] Furthermore, the parameters of the ship energy efficiency operation include the ship energy efficiency operation index EEOI, which is calculated as follows:

[0012]

[0013] Among them, FC M The total amount of fuel consumed by the main engine of the ship during navigation, FC A is the total fuel consumed by the auxiliary engine of the ship during navigation, C FM is the CO2 emission factor of the host, C FA is the auxiliary engine CO2 emission factor, which refers to the mass of CO2 emitted per ton of fuel consumed; m cargo is the load; D is the mileage of the ship;

[0014] When considering real-time navigation data, the relationship model between the main engine's ship energy efficiency operation index EEOIm and wind, waves, and speed is:

[0015]

[0016] Among them, FC j is the consumption of fuel j, C Fj is the carbon dioxide emission factor of fuel j, V is the ship speed, Indicates the speed V i The time taken to sail the distance Di, C1 is the sailing resistance parameter, C2 is the wave resistance parameter, and C3 is the wind resistance parameter.

[0017] Furthermore, C1, C2, and C3 are calculated as follows:

[0018]

[0019] Where ρ is the density of seawater, S is the wet surface area of the ship, and C t is the total resistance coefficient of the ship, SFC is the fuel consumption rate, PC is the ship propulsion coefficient, R wave is the wave resistance, R wind is wind resistance.

[0020] Further, the calculation method of the Ship Energy Efficiency Operational Index of the auxiliary engine is the same as that of the main engine.

[0021] Further, the specific analysis process of the AI model optimization analysis module is as follows:

[0022] Store the external real-time data into SQL and connect it to the AI basic framework; write the SPO optimization algorithm into the AI framework; the SPO optimization algorithm is improved based on the simulated annealing algorithm.

[0023] The SPO optimization algorithm includes the following steps:

[0024] (1) Algorithm initialization:

[0025] (1.1) Set the SPO algorithm parameters, including the number of SPO particles n, the maximum number of iterations X, the inertia weight w and the inertia weight w decreases linearly during the iteration, the learning factors z1 and z2, the SPO second-stage trigger interval, and the SPO second-stage initial temperature T0 and the cooling rate;

[0026] Set the objective function:

[0027]

[0028] (1.2) Define the variable boundaries, including the speed boundary [Vmin, Vmax], the cargo volume boundary [Loadmin, Loadmax], and the main engine efficiency boundary [Effmin, Effmax];

[0029] (1.3) Initialize the particle swarm, each particle randomly generates an initial position P0, the initial velocity Vi of particle i is 0, and record the individual best pBest and the global best gBest;

[0030] (2) Execute the SPO first-stage iteration:

[0031] (2.1) Traverse the particle swarm to calculate the current EEOIm, and calculate the fitness;

[0032] (2.2) Update the individual optimal solution, if the current EEOIm < pBest EEOIm, then update pBest;

[0033] (2.3) Update the global optimal solution, if the current EEOIm < gBest EEOIm, then update gBest;

[0034] (3) Trigger the SPO second-stage enhancement stage:

[0035] (3.1) Perform the second phase of SPO optimization at the trigger interval, apply random perturbations to the gBest position to generate a neighborhood solution; use the acceptance criterion to accept or reject the new solution;

[0036] (3.2) Update the global optimal solution. If the optimization result of the second stage of SPO is better than gBest, update gBest.

[0037] (3.3) Information is fed back to the particle swarm, and a particle is randomly selected to replace the optimization result of the second stage of SPO;

[0038] (4) Update particle state: velocity update, according to the formula Vn = w × Vi + × z1 (pBest - Pi) + z2 × (gBest - Pi), where Vn is the new velocity and Vi is the old velocity; position update, calculate the new position Pn = Pi + Vn, Pn is the new position, Pi is the old position; boundary constraint processing, prevent variables from exceeding the set range;

[0039] (5) Output optimization results: After the iteration, the global optimal solution gBest is returned, and the optimal speed and cargo volume m are analyzed and output. cargo , EEOI value.

[0040] Furthermore, the acceptance criterion is specifically that if the optimized EEOIm value is smaller than the old EEOIm value, it is accepted; if the optimized EEOIm value is larger than the old EEOIm value, the optimized value is likely accepted, and the acceptance probability is calculated by -ΔE / T, where ΔE is the difference between the values before and after optimization of EEOIm, and T is the current temperature.

[0041] Furthermore, the random perturbation generated by the gBest position is as follows: speed ±0.5 knots, cargo volume m cargo ±5%, EEOIm value ±0.05.

[0042] Furthermore, the boundary constraint processing is specifically as follows: limiting the variable values of the updated position P to their corresponding boundary ranges, that is, if they exceed the speed boundary, cargo volume boundary or main engine efficiency boundary, their corresponding values are adjusted to the corresponding boundary values.

[0043] Furthermore, the system also includes: an operation display module, which displays in real time the control operations of the AI autonomous operation module on the ship's main engine and auxiliary engine, and displays the current ship operation status, including speed, wave resistance, and wind resistance.

[0044] A method for optimizing EEOI based on an AI-simulated navigation data model includes the following steps:

[0045] (1) Ship sensors and monitoring equipment to collect ship fuel consumption data, energy efficiency parameters and external environment data in real time;

[0046] (2) The AI model optimization analysis module calculates the parameters of ship energy efficiency operation by processing real-time data information, and generates optimization suggestions for ship energy efficiency operation based on the optimization analysis results;

[0047] (3) After receiving the optimization suggestions for ship energy efficiency operation, the AI autonomous operation module actively intervenes to control the ship's main engine speed and the output of the ship's energy consumption devices. Then, the real-time navigation data according to the optimization suggestions is collected again and continuously transmitted to the AI model optimization analysis module for further optimization. The optimized data is then continuously submitted to the AI autonomous operation module.

[0048] (4) When the ship encounters an emergency, it will prompt human intervention.

[0049] Beneficial effects: The present invention introduces EEOI (Ship Energy Efficiency Operation Index) as an optimization indicator. The system introduces an AI model to calculate the optimal speed based on fuel consumption and carbon dioxide emissions at different speeds, so that the ship can significantly reduce fuel consumption while ensuring transportation efficiency and improve the economic benefits of the ship. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0051] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0052] like Figure 1 As shown, the present invention provides a technical solution:

[0053] A system for optimizing EEOI (Energy Output Index) based on an AI-powered simulated navigation data model includes a data acquisition module, an AI model optimization and analysis module, an AI autonomous operation module, and a user active intervention module. The data acquisition module is connected to ship sensors and monitoring equipment to collect real-time data on the ship's fuel consumption, energy efficiency parameters, and external environmental data. EEOI is a ship energy efficiency evaluation indicator that can more directly represent the amount of CO2 emitted during ship transportation. A smaller value indicates a higher energy efficiency of the ship.

[0054] The AI model optimization and analysis module calculates the parameters of ship energy efficiency operation by processing real-time data information, and generates optimization suggestions for ship energy efficiency operation based on the optimization analysis results;

[0055] After receiving the optimization suggestions for ship energy efficiency operation, the AI autonomous operation module actively intervenes to control the ship's main engine speed and the output of the ship's energy consumption devices. It then collects real-time navigation data based on the optimization suggestions and continues to pass it to the AI model optimization analysis module for further optimization. The optimized data is then submitted to the AI autonomous operation module.

[0056] The user active intervention module will prompt manual intervention when the ship encounters an emergency.

[0057] Furthermore, the parameters of the ship energy efficiency operation include the ship energy efficiency operation index EEOI, which is calculated as follows:

[0058]

[0059] Among them, FC M The total amount of fuel consumed by the main engine of the ship during navigation, FC A is the total fuel consumed by the auxiliary engine of the ship during navigation, C FM is the CO2 emission factor of the host, C FA is the auxiliary engine CO2 emission factor, which refers to the mass of CO2 emitted per ton of fuel consumed; m cargo is the load; D is the mileage of the ship;

[0060] When considering real-time navigation data, the relationship model between the main engine's ship energy efficiency operation index EEOIm and wind, waves, and speed is:

[0061]

[0062] Among them, FC j is the consumption of fuel j, C Fj is the carbon dioxide emission factor of fuel j, V is the ship speed, Indicates the speed V i The time taken to sail the distance Di, C1 is the sailing resistance parameter, C2 is the wave resistance parameter, and C3 is the wind resistance parameter.

[0063] Furthermore, C1, C2, and C3 are calculated as follows:

[0064]

[0065] Where ρ is the density of seawater, S is the wet surface area of the ship, and C t is the total resistance coefficient of the ship, SFC is the fuel consumption rate, PC is the ship propulsion coefficient, R wave is the wave resistance, R wind is wind resistance.

[0066] The wave resistance formula is as follows:

[0067]

[0068] Where, V represents the ship speed, m / s; B represents the ship width, m; d represents the ship draft, Lp represents the length between the ship's perpendiculars; hs represents the significant wave height, m.

[0069] The formula for wind resistance is as follows:

[0070]

[0071] Where Ca represents the air resistance coefficient, A T It represents the projected area of the hull and superstructure above the waterline on the mid-transverse section, ρa represents the air density, and Va represents the relative wind speed.

[0072] Furthermore, the calculation method of the ship energy efficiency operation index of the auxiliary engine is the same as that of the main engine.

[0073] Furthermore, the specific analysis process of the AI model optimization analysis module is as follows:

[0074] Store external real-time data into SQL and connect it to the AI infrastructure framework; write the SPO optimization algorithm into the AI framework;

[0075] The SPO optimization algorithm includes the following steps:

[0076] (1) Algorithm initialization:

[0077] (1.1) Setting the SPO algorithm parameters, including the number of SPO particles n, the maximum number of iterations X, the inertia weight w, which decreases linearly during the iteration, the learning factors z1 and z2, the SPO second stage trigger interval (the smaller the value, the more frequent the local search), and the initial temperature T0 and cooling rate of the SPO second stage;

[0078] Set the objective function:

[0079]

[0080] (1.2) Define the variable boundaries, including the speed boundary [Vmin, Vmax], the cargo volume boundary [Loadmin, Loadmax], and the main engine efficiency boundary [Effmin, Effmax];

[0081] (1.3) Initialize the particle swarm, randomly generate an initial position P0 for each particle, initialize the velocity Vi of particle i to 0, and record the individual optimal pBest and global optimal gBest;

[0082] (2) Execute the first phase of SPO iteration:

[0083] (2.1) Traverse the particle swarm to calculate the current EEOIm and calculate the fitness;

[0084] (2.2) Update the individual optimal solution. If the current EEOIm < pBest EEOIm, then update pBest;

[0085] (2.3) Update the global optimal solution. If the current EEOIm < gBest EEOIm, then update gBest;

[0086] (3) Trigger the second stage (enhancement stage) of SPO:

[0087] (3.1) Execute the second stage optimization of SPO at the trigger interval, generate a neighborhood solution by applying a random perturbation to the above gBest position; use the acceptance criterion to accept or reject the new solution;

[0088] (3.2) Update the global optimal solution. If the optimization result of the second stage of SPO is better than gBest, update gBest;

[0089] (3.3) Feed back information to the particle swarm, randomly select a particle and replace it with the optimization result of the second stage of SPO;

[0090] (4) Update the particle state: Update the velocity according to the formula Vn = w×Vi + z1×(pBest - Pi) + z2×(gBest - Pi), where Vn is the new velocity and Vi is the old velocity; Update the position, calculate the new position Pn = Pi + Vn, where Pn is the new position and Pi is the old position; Process boundary constraints to prevent variables from exceeding the set range;

[0091] (5) Output the optimization result: After the iteration ends, return the global optimal solution gBest, and parse and output the optimal ship speed and cargo volume m cargo and the EEOI value.

[0092] Further, the acceptance criterion is specifically that if the value of EEOIm after optimization is smaller than the old value of EEOIm, it is accepted; if the value of EEOIm after optimization is larger than the old value of EEOIm, the value after optimization is accepted with a probability, and the acceptance probability is calculated by -ΔE / T, where ΔE is the difference in the value of EEOIm before and after optimization, and T is the current temperature.

[0093] Further, the generation of the neighborhood solution by applying a random perturbation to the above gBest position is specifically: ship speed ±0.5 knots, cargo volume m cargo ±5%, EEOIm value ±0.05.

[0094] Furthermore, the boundary constraint processing is specifically as follows: limiting the variable values of the updated position P to their corresponding boundary ranges, that is, if they exceed the speed boundary, cargo volume boundary or main engine efficiency boundary, their corresponding values are adjusted to the corresponding boundary values.

[0095] Furthermore, the system also includes: an operation display module, which displays in real time the control operations of the AI autonomous operation module on the ship's main engine and auxiliary engine, and displays the current ship operation status, including speed, wave resistance, and wind resistance.

[0096] A method for optimizing EEOI based on an AI-simulated navigation data model includes the following steps:

[0097] (1) Ship sensors and monitoring equipment to collect ship fuel consumption data, energy efficiency parameters and external environment data in real time;

[0098] (2) The AI model optimization analysis module calculates the parameters of ship energy efficiency operation by processing real-time data information, and generates optimization suggestions for ship energy efficiency operation based on the optimization analysis results;

[0099] (3) After receiving the optimization suggestions for ship energy efficiency operation, the AI autonomous operation module actively intervenes to control the ship's main engine speed and the output of the ship's energy consumption devices. Then, the real-time navigation data according to the optimization suggestions is collected again and continuously transmitted to the AI model optimization analysis module for further optimization. The optimized data is then continuously submitted to the AI autonomous operation module.

[0100] (4) When the ship encounters an emergency, it will prompt human intervention.

Claims

1. A system for optimizing EEOI based on AI simulation navigation data model, characterized by: It includes a data acquisition module, an AI model optimization and analysis module, an AI autonomous operation module, and a user active intervention module; the data acquisition module is connected to ship sensors and monitoring equipment to collect real-time fuel consumption data, energy efficiency parameters, and external environment data of the ship; The AI model optimization and analysis module calculates the parameters of ship energy efficiency operation by processing real-time data information, and generates optimization suggestions for ship energy efficiency operation through the optimization analysis results; After receiving the optimization suggestions for ship energy efficiency operation, the AI autonomous operation module actively intervenes to control the rotation speed of the ship's main engine and the output of on-board energy-consuming devices, then collects the real-time navigation data after following the optimization suggestions and continues to transmit it to the AI model optimization and analysis module for further optimization; and the optimized data is continuously submitted to the AI autonomous operation module; The user active intervention module will remind manual intervention when the ship encounters emergencies.

2. The system for optimizing EEOI based on AI simulated navigation data model according to claim 1, characterized in that: The parameters of ship energy efficiency operation include the ship energy efficiency operation index EEOI, and the formula is as follows: Among them, FC M The total amount of fuel consumed by the main engine of the ship during navigation, FC A is the total fuel consumed by the auxiliary engine of the ship during navigation, C FM is the CO2 emission factor of the host, C FA is the auxiliary engine CO2 emission factor, which refers to the mass of CO2 emitted per ton of fuel consumed; m cargo is the load; D is the mileage of the ship; When considering real-time navigation data, the relationship model between the ship energy efficiency operation index EEOIm of the main engine and wind waves and ship speed: Among them, FC j is the consumption of fuel j, C Fj is the carbon dioxide emission factor of fuel j, V is the ship speed, Indicates the speed V i The time taken to sail the distance Di, C1 is the sailing resistance parameter, C2 is the wave resistance parameter, and C3 is the wind resistance parameter.

3. The system for optimizing EEOI based on AI simulated navigation data model according to claim 2, characterized in that: C1, C2, and C3 are specifically calculated as follows: Where ρ is the density of seawater, S is the wet surface area of the ship, and C t is the total resistance coefficient of the ship, SFC is the fuel consumption rate, PC is the ship propulsion coefficient, R wave is the wave resistance, R wind is wind resistance.

4. The system for optimizing EEOI based on AI simulated navigation data model according to claim 2, characterized in that: The calculation method of the ship energy efficiency operation index of the auxiliary engine is the same as that of the main engine.

5. The system for optimizing EEOI based on AI simulated navigation data model according to claim 2, characterized in that: The specific analysis process of the AI model optimization and analysis module is as follows: Store external real-time data in SQL and connect it to the AI basic framework; write the SPO optimization algorithm into the AI framework; The SPO optimization algorithm includes the following steps: (1) Algorithm initialization: (1.1) Set the SPO algorithm parameters, including the number of SPO particles n, the maximum number of iterations X, the inertia weight w (and the inertia weight w linearly decreases during the iteration), the learning factors z1 and z2, the trigger interval of the second stage of SPO, and the initial temperature T0 and cooling rate of the second stage of SPO; Set the objective function: (1.2) Define the variable boundaries, including the ship speed boundary [Vmin, Vmax], the cargo volume boundary [Loadmin, Loadmax], and the main engine efficiency boundary [Effmin, Effmax]; (1.3) Initialize the particle swarm, randomly generate the initial position P0 for each particle, initialize the velocity Vi of particle i to 0, and record the individual best pBest and the global best gBest; (2) Execute the first stage iteration of SPO: (2.1) Traverse the particle swarm to calculate the current EEOIm and calculate the fitness; (2.2) Update the individual optimal solution. If the current EEOIm < pBest EEOIm, then update pBest; (2.3) Update the global optimal solution. If the current EEOIm < gBest EEOIm, then update gBest; (3) Trigger the second stage enhancement stage of SPO: (3.1) Execute the second stage optimization of SPO at the trigger interval, generate a neighborhood solution by applying a random perturbation to the above gBest position; use the acceptance criterion to accept or reject the new solution; (3.2) Update the global optimal solution. If the optimization result of the second stage of SPO is better than gBest, update gBest; (3.3) Information is fed back to the particle swarm, and a particle is randomly selected to replace the optimization result of the second stage of SPO; (4) Update particle state: velocity update, according to the formula Vn = w × Vi + × z1 (pBest - Pi) + z2 × (gBest - Pi), where Vn is the new velocity and Vi is the old velocity; position update, calculate the new position Pn = Pi + Vn, Pn is the new position, Pi is the old position; boundary constraint processing, prevent variables from exceeding the set range; (5) Output optimization results: After the iteration, the global optimal solution gBest is returned, and the optimal speed and cargo volume m are analyzed and output. cargo , EEOI value.

6. The system for optimizing EEOI based on AI simulated navigation data model according to claim 5, characterized in that: The acceptance criteria are as follows: if the optimized EEOIm value is smaller than the old EEOIm value, it is accepted; if the optimized EEOIm value is larger than the old EEOIm value, the optimized value is likely accepted. The acceptance probability is calculated by -ΔE / T, where ΔE is the difference between the values before and after optimization of EEOIm, and T is the current temperature.

7. The system for optimizing EEOI based on AI simulated navigation data model according to claim 5, characterized in that: The random perturbation generated based on the gBest position is as follows: speed ±0.5 knots, cargo volume m cargo ±5%, EEOIm value ±0.

05.

8. The system for optimizing EEOI based on AI simulated navigation data model according to claim 5, characterized in that: The boundary constraint processing is specifically as follows: limiting the variable values of the updated position P to their corresponding boundary ranges, that is, if they exceed the speed boundary, cargo volume boundary or host efficiency boundary, their corresponding values are adjusted to the corresponding boundary values.

9. The system for optimizing EEOI based on AI simulated navigation data model according to claim 1, characterized in that: The system also includes: an operation display module, which displays in real time the control operations of the AI autonomous operation module on the ship's main engine and auxiliary engine, and displays the current ship operation status, including speed, wave resistance, and wind resistance.

10. A method for optimizing EEOI based on AI simulation navigation data model, characterized in that: The following steps are involved: (1) Ship sensors and monitoring equipment to collect ship fuel consumption data, energy efficiency parameters and external environment data in real time; (2) The AI model optimization analysis module calculates the parameters of ship energy efficiency operation by processing real-time data information, and generates optimization suggestions for ship energy efficiency operation based on the optimization analysis results; (3) After receiving the optimization suggestions for ship energy efficiency operation, the AI autonomous operation module actively intervenes to control the ship's main engine speed and the output of the ship's energy consumption devices. Then, the real-time navigation data according to the optimization suggestions is collected again and continuously transmitted to the AI model optimization analysis module for further optimization. The optimized data is then continuously submitted to the AI autonomous operation module. (4) When the ship encounters an emergency, it will prompt human intervention.

Images