A ship speed optimization method with self-learning ability

Abstract

The present invention discloses a ship speed optimization method with self-learning ability. First, an initial relationship model between fuel consumption and speed is established. Then, after learning and optimization, the draft of the ship is obtained according to the actual state of the ship floating in the water during actual operation. The mathematical model is solved with the minimum energy consumption as the optimization goal to obtain the speed. The present invention combines the ship design theory, takes into account the influence of draft and wind and waves on the speed, and makes the speed fuel consumption model more accurate; the basic mathematical model of speed fuel consumption of the present invention is established based on the ship's own speed power curve and the main engine's power fuel consumption characteristic curve. The model is established on a correct theoretical basis, and the system can provide speed optimization auxiliary decision-making after being installed on board. The present invention has a basic mathematical model, which can correct the actual fuel consumption to an ideal environment and compare it with the data in the basic model, so as to assist the shipowner in determining the bottom fouling and make auxiliary decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of ship design, and more particularly to the optimization of ship speed. Background Art

[0002] The environmental awareness of the international community is increasing day by day, and the requirements of the shipping industry for environmental protection are also increasing. In recent years, SEEMP, EEDI, EEXI and EEOI have been implemented successively, aiming to urge ships to save energy and reduce emissions, improve energy efficiency management, and reduce greenhouse gas emissions. In order to improve the energy efficiency of ships, reduce fuel consumption and reduce carbon dioxide emissions, shipping companies use sails, air lubrication and other equipment on ships to reduce ship energy consumption and improve ship operation efficiency.

[0003] Under the development trend of green intelligence, green means alone are not enough to achieve the best energy-saving effect. It is necessary to effectively use intelligent means based on green technology to improve the efficiency of ship energy management and reduce ship fuel consumption and carbon dioxide emissions.

[0004] At present, some intelligent energy efficiency management systems on the market have the function of optimizing ship speed, but their speed optimization has limitations. The main problems are as follows:

[0005] 1. The speed optimization in the current energy efficiency system is based on the statistics and analysis of the fuel consumption and speed of the main and auxiliary engines collected during the operation of the ship to establish a mathematical model of speed and fuel consumption, and then propose auxiliary decision-making for speed optimization. However, the established model is separated from the speed and fuel consumption of the ship itself;

[0006] 2. The mathematical model of ship speed and fuel consumption is also related to the wind and wave conditions. The current energy efficiency system model does not consider the effects of draft and buoyancy;

[0007] 3. The speed optimization function of the current energy efficiency system needs to be used for a long time before it can give reasonable speed optimization suggestions. Summary of the invention

[0008] In view of the above problems, the present invention provides a ship speed optimization method with self-learning ability. The speed and fuel consumption of the present invention establish an initial mathematical model based on ship model test results or numerical simulation calculation results; the speed and fuel consumption mathematical model takes into account the influence of wind and waves; the present invention can provide more reasonable speed optimization auxiliary decision-making after being installed on the ship. At the same time, the data model can continuously learn and optimize autonomously according to actual operating data, so as to improve the accuracy and rationality of speed optimization.

[0009] In order to achieve the above object, the present invention provides a ship speed optimization method with self-learning ability, comprising the following steps:

[0010] S1. Data collection:

[0011] g. Collect the ship's main engine power P ture ;

[0012] h. Collect the fuel consumption FOC of the ship's main engine ture ;

[0013] i. Collect the ship's speed over the ground V G

[0014] j. Collect the draft T of the bow and stern of the ship F / T A

[0015] k. Collect the relative wind speed V around the ship WRref and relative wind direction ψ WRref ;

[0016] l. The significant wave height H of the collected waves 1 / 3 ;

[0017] S2. Establish the initial relationship model between fuel consumption FOC and speed V, including the following steps

[0018] 1) A scaled model is established according to the design line of the ship. The actual state of the ship sailing in the seawater is simulated by sailing in the test pool. The relationship between the main engine power P and the speed V can be obtained by measurement. The specific process is as follows:

[0019] e) Under the selected ship draft T1, the ship sails at a speed V1 in the test water tank, and the ship's resistance R1 and propulsion efficiency η1 are measured, and the corresponding main engine power P1=R1×η1 is calculated; the ship sails at a speed V2 in the test water tank, and the ship's resistance R2 and propulsion efficiency η2 are measured, and the corresponding main engine power P2=R2×η2……;

[0020] In the test tank, the ship speed V n Sailing, measuring the ship's resistance R n and propulsion efficiency η n , calculate the corresponding host power P n =R n ×η n , n is a natural number; by fitting the above series of data points, the relationship between propulsion power and ship speed at draft T1 can be obtained: P = f1 (V);

[0021] f) Repeat the process in a) at the selected ship draft T2, P = f2 (V);

[0022] g) At the selected ship draft T n Repeat the process a), P = f n (V);

[0023] h) By fitting a series of relationships between draft and propulsion power obtained from a) to c), the relationship between propulsion power P and ship speed V considering the draft factor can be obtained, P = f(V, T);

[0024] 2) When the host power is P n When the fuel consumption per kilowatt of main engine power is recorded as SFOC, the main engine power P n Fuel consumed FOC = P n ×SFOC;

[0025] 3) Combine 1) and 2) to establish the initial relationship model between fuel consumption FOC and speed V: FOC = SFOC × f (V, T);

[0026] S3, analysis and processing, including the following steps:

[0027] d. Wind resistance factor

[0028] The wind in the ocean will cause the resistance of the ship to increase. The increased resistance R AA =0.5ρ A ·C AA (ψ WRref )·A XV ·V WRref 2 -0.5ρ A ·C AA (0)·A XV 2 ·V G 2

[0029] Where: R AA Indicates the increase in resistance caused by wind;

[0030] ρ A is the density of air;

[0031] C AA (ψ WRref ) is the wind resistance coefficient in the direction relative to the wind;

[0032] A XV is the windward area;

[0033] C AA (0) is the wind resistance coefficient against the wind;

[0034] V G is the ship’s speed over ground;

[0035] Therefore, the wind resistance factor can be expressed as the mathematical model f wind =f(ψ WRref ,VWRref );

[0036] When the ship is in operation, the relative wind direction ψ is collected daily. WRref-n and relative wind speed V WRref-n , calculate the wind resistance influence factor f wind-n , gradually improve the wind resistance influencing factor model f wind =f(ψ WRref ,V WRref ), gradually approaching the real situation;

[0037] e. Wave resistance influence factor

[0038] Waves in the ocean will increase the resistance a ship experiences.

[0039] Where: R AWL Indicates the increase in resistance caused by wind;

[0040] ρ S is the density of seawater;

[0041] g is the acceleration due to gravity;

[0042] H 1 / 3 For the meaningful height of the sea water;

[0043] B is the ship's breadth;

[0044] L BWL For the captain;

[0045] Therefore, the wave resistance factor can be expressed as f wave =f(H 1 / 3 );

[0046] During the operation of the ship, the significant wave height H is collected daily. 1 / 3-n , calculate the wave resistance factor f wave-n ; Gradually improve the wave resistance influencing factor model f wave =f(H 1 / 3 ), gradually approaching the real impact situation;

[0047] f. Fuel consumption difference factor

[0048] SFOC in actual operation ture The main engine power P collected by the shaft power meter ture Fuel consumption FOC collected by mass flow meter ture Calculated,

[0049] Introducing fuel consumption difference factor Therefore, the fuel consumption difference factor can be expressed as f SFOC =f(Pture ,FOC ture );

[0050] When the ship is in operation, the main engine power P is collected daily. ture-n and fuel consumption FOC ture-n , calculate the wave resistance influence factor f SFOC-n ; Gradually improve the fuel consumption difference factor model f SFOC =f(P ture ,FOC ture ), gradually approaching the real situation;

[0051] S4, speed optimization, including the following steps:

[0052] c. According to steps S2 and S3, the relationship model between the fuel consumption FOC and the speed V is optimized to FOC = f wind ×f wave ×f SFOC ×SFOC×f(V,T);

[0053] d. The distance the ship travels every day is S = V × 24,

[0054] The fuel consumption per unit distance is

[0055] In the actual operation process, the draft T of the ship is obtained according to the actual state of the ship floating in the water, and the mathematical model is solved with the minimum FC as the optimization goal, and the speed V is obtained. O .

[0056] In the preferred method, the main engine power P of the ship is collected by setting a shaft power meter ture , host power P ture Refers to the amount of power provided by the main engine to rotate the propeller and propel the ship forward. The fuel consumption FOC of the ship's main engine is collected by installing a mass flow meter on the main engine supply system pipeline. ture The ship’s speed over the ground V is collected through the global satellite positioning system (GPS). G The draft measurement equipment is used to collect the draft T of the bow and stern of the ship. F / T A The wind speed and direction instrument is installed on the top of the ship to collect the relative wind speed V around the ship. WRref and relative wind direction ψ WRref The significant wave height H of the wave is measured by wave radar 1 / 3 .

[0057] In order to achieve the above-mentioned purpose, the present invention combines the ship design theory, takes into account the influence of draft and wind and waves on the speed, and makes the speed fuel consumption model more accurate; the basic mathematical model of speed fuel consumption of the present invention is established based on the ship's own speed power curve and the main engine's power fuel consumption characteristic curve. The model is established on a correct theoretical basis, and the system can provide auxiliary decision-making for speed optimization after being installed on board. With the data collected during the operation of the present invention, the speed fuel consumption model will continue to learn autonomously, so that the model gradually approaches the actual energy consumption of the ship's operation, and improves the accuracy of auxiliary decision-making for speed optimization; the present invention has a basic mathematical model, which can correct the actual fuel consumption to an ideal environment and compare it with the data in the basic model, so as to help shipowners determine the bottom of the fouling and make auxiliary decisions. DETAILED DESCRIPTION

[0058] The present invention provides a ship speed optimization method with self-learning ability, which can be implemented through a system, wherein the system consists of four parts: a data acquisition module, a data analysis module, a ship database module and a speed optimization auxiliary decision module, wherein the data acquisition module is a hardware module, and the database module, the data analysis module and the speed optimization auxiliary decision module are software modules.

[0059] The data acquisition module is used to collect various data of the ship itself and the marine environment for data analysis by a ship speed optimization system with self-learning ability, including shaft power meter, mass flow meter, wind speed and direction meter, wave radar, global satellite positioning system (GPS), speed meter, ship telemeter and other measuring instruments:

[0060] The shaft power device collects the ship's main engine power P ture , host power P ture Refers to the power provided by the main engine to rotate the propeller and propel the ship forward;

[0061] The mass flow meter is installed on the main engine supply system pipeline to collect the fuel consumption FOC of the ship's main engine. ture ;

[0062] The global satellite positioning system (GPS) collects the ship's ground speed V G

[0063] The draft measuring equipment collects the draft T of the bow and stern of the ship. F / T A

[0064] The wind speed and direction instrument is installed on the top of the ship to measure the relative wind speed V around the ship. WRref and relative wind direction ψ WRref ;

[0065] Wave radar measures the significant wave height H of waves 1 / 3 ;

[0066] The database is the basis for the data analysis module to analyze navigation data, and also the basis for the speed optimization auxiliary decision module to provide the best speed recommendation. The database stores a large amount of data, including the initial relationship model between fuel consumption FOC and speed V and the relationship model between fuel consumption FOC and speed V after learning and optimization.

[0067] The initial fuel consumption FOC and speed V relationship model is established based on the model test results, including the following steps:

[0068] 1. A scaled model of the ship is established based on the design line or outline of the ship. The navigation of the scaled model of the ship in the test pool simulates the state of the actual ship sailing in the seawater. The relationship between the main engine power P and the speed V can be obtained through measurement.

[0069] a. Under the selected draft T1, the ship sails at a speed V1 in the test pool, and the resistance R1 and propulsion efficiency η1 of the ship are measured, and the corresponding main engine power P1=R1×η1 is calculated; when the ship sails at a speed V2 in the test pool, the resistance R2 and propulsion efficiency η2 of the ship are measured, and the corresponding main engine power P2=R2×η2 is calculated; when the ship sails at a speed V n Sailing, measuring the ship's resistance R n and propulsion efficiency η n , calculate the corresponding host power P n =R n ×η n By fitting the above series of data points, we can get the relationship between propulsion power and ship speed at draft T1: P = f1 (V);

[0070] b. Repeat the process in a) at the selected ship draft T2, P = f2 (V);

[0071] c. At the selected ship draft T n Repeat the process a), P = f n (V);

[0072] By fitting a series of relationships between draft and propulsion power obtained from a) to c), the relationship between propulsion power P and speed V considering the draft factor can be obtained, P = f (V, T);

[0073] 2. According to the factory test report of the main engine, the fuel consumption of the main engine under different loads can be obtained: when the main engine power is P n When the fuel consumption per kilowatt of main engine power is recorded as SFOC, the main engine power P n Fuel consumed FOC = P n ×SFOC;

[0074] 3. Combine 1) and 2) to establish the initial relationship model between fuel consumption FOC and speed V: FOC = SFOC × f (V, T)

[0075] Afterwards, the data analysis module analyzes and processes the collected data.

[0076] a. Wind resistance influencing factors

[0077] The "ISO 15016-2015 Ships and marine technology. Guidelines for the assessment of speed and power performance by analysis of speed test data" issued by the International Organization for Standardization ISO proposes the influence of wind and wave factors on the relationship between the ship's main engine power P and speed V, P = f(V, T).

[0078] The wind in the ocean will increase the resistance of the ship. According to the requirements of ISO 15016-2015, the increased resistance R AA =0.5ρ A ·C AA (ψ WRref )·A XV ·V WRref 2 -0.5ρ A ·C AA (0)·A XV 2 ·V G 2

[0079] Where: R AA Indicates the increase in resistance caused by wind;

[0080] ρ A is the density of air, which is based on the standard value of ISO 15016-2015;

[0081] C AA (ψ WRref ) is the wind resistance coefficient in the relative wind direction, which can be calculated based on the measured relative wind direction ψ WRref Check it out in ISO15016-2015;

[0082] A XV is the windward area, which can be obtained by consulting the ship's design drawings;

[0083] C AA (0) is the wind resistance coefficient against wind, obtained according to ISO 15016-2015;

[0084] V G is the ship’s speed over the ground, measured by the Global Positioning System (GPS);

[0085] This data analysis module introduces wind resistance influencing factors According to the above formula, it can be found that wind resistance is related to relative wind direction ψ WRref and relative wind speed V WRref Therefore, the wind resistance influencing factor can be expressed as the mathematical model f wind =f(ψ WRref ,V WRref ).

[0086] When the ship is in operation for one day, the relative wind direction ψ WRref-n and relative wind speed V WRref-n , calculate the wind resistance influence factor f wind-n . Collect a set of data every day and perform corresponding calculations, add a set of data to the wind resistance coefficient model, and improve the wind resistance influencing factor model f wind =f(ψ WRref ,V WRref ), continuously learns and evolves autonomously during operation, making the mathematical model of wind resistance influencing factors f w =f(ψ WRref ,V WRref ) gradually approaches the actual impact situation.

[0087] b. Wave resistance influence factor

[0088] Waves in the ocean will increase the resistance of the ship. According to ISO 15016-2015, the increased resistance

[0089] Where: R AWL Indicates the increase in resistance caused by wind;

[0090] ρ S is the density of seawater, which is the standard value according to ISO 15016-2015;

[0091] g is the acceleration due to gravity, which is 9.8m / m2;

[0092] H 1 / 3 is the significant wave height of seawater, measured by wave radar;

[0093] B is the ship's breadth, obtained from the ship's design drawings (general arrangement drawings);

[0094] L BWL is the captain, obtained from the ship's design drawings (general arrangement drawings);

[0095] This data analysis module introduces the wave resistance influencing factor According to the above formula, it can be found that the wave resistance is related to the significant wave height H 1 / 3 Therefore, the wave resistance factor can be expressed as fwave =f(H 1 / 3 ).

[0096] When the ship is in operation for one day, the significant wave height H is collected. 1 / 3-n , calculate the wave resistance factor f wave-n A set of data is collected every day and corresponding calculations are performed to add a set of data to the wave resistance coefficient model and improve the wave resistance influencing factor model f wave =f(H 1 / 3 ), continuously learns and evolves autonomously during operation, making the mathematical model of wave resistance influencing factors f wave =f(H 1 / 3 ) gradually approaches the actual impact situation.

[0097] c. Fuel consumption difference factor

[0098] Since the SFOC in the above article is obtained through the factory test report, the SFOC in the actual operation process is ture There will be some differences between the SFOC tested in the factory and the SFOC in the actual operation process. ture The main engine power P can be collected by the shaft power meter ture Fuel consumption FOC collected by mass flow meter ture Calculated,

[0099] This model introduces the fuel consumption difference factor According to the above formula, it can be found that the difference factor is related to the host power P ture and fuel consumption FOC ture Therefore, the fuel consumption difference factor can be expressed as f SFOC =f(P ture ,FOC ture ).

[0100] When the ship is in operation for one day, the main engine power P is collected. ture-n and fuel consumption FOC ture-n , calculate the wave resistance influence factor f SFOC-n . Collect a set of data every day and perform corresponding calculations, add a set of data to the fuel consumption difference factor model, and improve the fuel consumption difference factor model f SFOC =f(P ture ,FOC ture ), continuously learns and evolves autonomously during operation, making the mathematical model of fuel consumption difference factor f SFOC =f(P ture ,FOC ture ) gradually approaches the actual impact situation.

[0101] Finally, the speed optimization auxiliary decision module is a module that provides crew members with optimized speed auxiliary decision-making with the lowest fuel consumption.

[0102] According to the above, the relationship model between the initial fuel consumption FOC and the speed V is established as FOC = SFOC × f (V, T); Claim (3) obtains the wind resistance influencing factor f based on the data analysis of the actual operation process of the ship wind , wave resistance factor f wave , fuel consumption difference factor f SFOC Considering the three influencing factors in actual operation, the relationship model between fuel consumption FOC and speed V is optimized as FOC = f wind ×f wave ×f SFOC ×SFOC×f(V,T);

[0103] The distance a ship travels every day is S = V × 24, that is, the distance a ship travels in 24 hours. The fuel consumption per unit distance is Therefore The final model of this speed optimization system;

[0104] In the actual operation process, the crew can obtain the draft T of the ship according to the actual state of the ship floating in the water, input the draft T into the speed optimization auxiliary decision module, solve the mathematical model with the minimum FC as the optimization goal, and obtain the speed V O That is the optimal speed V O , the ship saves the most fuel when running at this speed.

[0105] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A ship speed optimization method with self-learning ability, characterized in that: The steps include: S1. Data collection: a. Collect the ship's main engine power P ture ; b. Collect the fuel consumption FOC of the ship's main engine ture ; c. Collect the ship's speed over the ground V G d. Collect the draft T of the bow and stern of the ship F / T A e. Collect the relative wind speed V around the ship WRref and relative wind direction ψ WRref ; f. The significant wave height H of the collected waves 1 / 3 ; S2. Establish the initial relationship model between fuel consumption FOC and speed V, including the following steps 1) A scaled model is established according to the design line of the ship. The actual state of the ship sailing in the seawater is simulated by sailing in the test pool. The relationship between the main engine power P and the speed V can be obtained by measurement. The specific process is as follows: a) Under the selected ship draft T1, the ship sails at a speed V1 in the test water tank, and the ship's resistance R1 and propulsion efficiency η1 are measured, and the corresponding main engine power P1=R1×η1 is calculated; when the ship sails at a speed V2 in the test water tank, the ship's resistance R2 and propulsion efficiency η2 are measured, and the corresponding main engine power P2=R2×η2…… In the test tank, the ship speed V n Navigate and measure the resistance R of the ship n and propulsion efficiency η n , calculate the corresponding host power P n =R n ×η n , n is a natural number; by fitting the above series of data points, the relationship between propulsion power and ship speed at draft T1 can be obtained: P = f1 (V); b) Repeat the process in a) at the selected ship draft T2, P = f2 (V); c) At the selected ship draft T n Repeat the process a), P = f n (V); d) By fitting a series of relationships between draft and propulsion power obtained from a) to c), the relationship between propulsion power P and ship speed V considering the draft factor can be obtained: P = f(V, T); 2) When the host power is P n When the fuel consumption per kilowatt of main engine power is recorded as SFOC, the main engine power P n Fuel consumed FOC = P n ×SFOC; 3) Combine 1) and 2) to establish the initial relationship model between fuel consumption FOC and speed V: FOC = SFOC × f (V, T); S3, analysis and processing, including the following steps: a. Wind resistance factors The wind in the ocean will cause the resistance of the ship to increase. The increased resistance R AA =0.5ρ A ·C AA (ψ WRref )·A XV ·V WRref 2 -0.5ρ A ·C AA (0)·A XV 2 ·V G 2 Where: R AA Indicates the increase in resistance caused by wind; ρ A is the density of air; C AA (ψ WRref ) is the wind resistance coefficient in the direction relative to the wind; A XV is the windward area; C AA (0) is the wind resistance coefficient against the wind; V G is the ship’s speed over ground; Therefore, the wind resistance factor can be expressed as the mathematical model f wind =f(ψ WRref ,V WRref ); When the ship is in operation, the relative wind direction ψ is collected daily. WRref-n and relative wind speed V WRref-n , calculate the wind resistance influence factor f wind-n , gradually improve the wind resistance influencing factor model f wind =f(ψ WRref ,V WRref ), gradually approaching the real situation; b. Wave resistance influence factor Waves in the ocean will increase the resistance a ship experiences. Where: R AWL Indicates the increase in resistance caused by wind; ρ S is the density of seawater; g is the acceleration due to gravity; H 1 / 3 For the meaningful height of the sea water; B is the ship's breadth; L BWL For the captain; Therefore, the wave resistance factor can be expressed as f wave =f(H 1 / 3 ); During the operation of the ship, the significant wave height H is collected daily. 1 / 3-n , calculate the wave resistance factor f wave-n ; Gradually improve the wave resistance influencing factor model f wave =f(H 1 / 3 ), gradually approaching the real impact situation; c. Fuel consumption difference factor SFOC in actual operation ture The main engine power P collected by the shaft power meter ture Fuel consumption FOC collected by mass flow meter ture Calculated, Introducing fuel consumption difference factor Therefore, the fuel consumption difference factor can be expressed as f SFOC =f(P ture ,FOC ture ); When the ship is in operation, the main engine power P is collected daily. ture-n and fuel consumption FOC ture-n , calculate the wave resistance influence factor f SFOC-n ; Gradually improve the fuel consumption difference factor model f SFOC =f(P ture ,FOC ture ), gradually approaching the real situation; S4, speed optimization, including the following steps: a. According to steps S2 and S3, the relationship model between the fuel consumption FOC and the speed V is optimized to FOC = f wind ×f wave ×f SFOC ×SFOC×f(V,T); b. The distance the ship travels every day S = V × 24, The fuel consumption per unit distance is In the actual operation process, the draft T of the ship is obtained according to the actual state of the ship floating in the water, and the mathematical model is solved with the minimum FC as the optimization goal, and the speed V is obtained. O .

2. According to claim 1, the ship speed optimization method with self-learning ability is characterized in that: The ship's main engine power P is collected by setting up a shaft power meter ture , host power P ture Refers to the amount of power provided by the main engine to rotate the propeller and propel the ship forward.

3. According to claim 1, the ship speed optimization method with self-learning ability is characterized in that: The fuel consumption FOC of the ship's main engine is collected by installing a mass flow meter on the main engine supply system pipeline. ture .

4. According to claim 1, the ship speed optimization method with self-learning ability is characterized in that: The ship's ground speed V is collected through the global satellite positioning system (GPS) G .

5. According to claim 1, the ship speed optimization method with self-learning ability is characterized in that: The draft T of the bow and stern of the ship is collected by draft measuring equipment. F / T A .

6. The ship speed optimization method with self-learning capability according to claim 1 is characterized in that: The wind speed and direction instrument is installed on the top of the ship to collect the relative wind speed V around the ship. WRref and relative wind direction ψ WRref .

7. The ship speed optimization method with self-learning capability according to claim 1 is characterized in that: The significant wave height H is measured by wave radar 1 / 3 .

Images