Layout optimization method for single sail of wind-assisted propulsion ship

Abstract

The invention relates to the technical field of wind power assisted propulsion ships, and provides a single sail layout optimization method for a wind power assisted propulsion ship, which adopts a method of combining an artificial neural network and a multi-target genetic algorithm, namely, adopts the artificial neural network to construct an agent surface model of a relationship between the saving rate of the host power and the single sail layout; a proxy plane model of the relation between the propulsion efficiency of the main propulsion device and the single sail layout; and by taking the proxy surface as the fitness, performing multi-objective optimization on the layout of the single sail by applying a multi-objective genetic algorithm, and obtaining the optimal spatial layout (namely position) of the single sail in a given deck space of a given ship, a given single sail, a given navigational speed and a given deck space. According to the invention, multi-objective optimization can be realized, that is, both the saving rate maximization of the power of the main engine and the propulsion efficiency maximization of the main propulsion device are considered; meanwhile, the method has high efficiency and can be used for energy efficiency improvement feasibility analysis and preliminary design of the wind-assisted propulsion ship.

Description

Technical Field

[0001] The invention relates to the technical field of wind-assisted propulsion ships, and in particular to a method for optimizing the layout of a single sail of a wind-assisted propulsion ship. Background Art

[0002] The wind-assisted propulsion system provides thrust for the ship under favorable wind direction, reduces the effective power while maintaining the speed, and thus reduces the power required by the main engine. Given a ship, a single sail, a given speed, and a given deck space, how to optimize the layout of a single sail, that is, how to determine the position of a single sail, so as to maximize the power saving rate of the main engine and maximize the propulsion efficiency of the main propulsion device, has become a concern in the design of wind-assisted propulsion ships.

[0003] The existing optimization scheme for the layout of a single sail of a wind-assisted propulsion ship mainly uses the enumeration method, that is, continuously changing the position of a single sail in the deck space available for arrangement in small steps, obtaining the law of change of the main engine power saving rate with the position of a single sail, and finding the optimal position of a single sail.

[0004] The disadvantages of the existing solutions are:

[0005] (1) It is only applicable to single-objective optimization, not to multi-objective optimization;

[0006] (ii) Under each layout of a single sail, the power saving rate of the main engine and the propulsion efficiency of the main propulsion device need to be obtained by solving the energy-saving effect evaluation model of wind-assisted propulsion ships. The energy-saving effect evaluation model contains a set of four-degree-of-freedom mechanical equilibrium equations, as well as the relationship between the hydrodynamic force / aerodynamic force on the ship and the navigation attitude (rudder angle, drift angle, heel angle) and effective thrust in the equations. The relationship between hydrodynamic force and aerodynamic force and navigation attitude and effective thrust can be obtained through empirical / semi-empirical formulas, or by three-dimensional CFD calculations or model tests. When the latter is adopted, if the existing scheme, that is, the enumeration method, is used for optimization, the efficiency is low and it is not suitable for the feasibility analysis and preliminary design of energy efficiency improvement of wind-assisted propulsion ships. Summary of the invention

[0007] In order to solve the technical problem that the existing technology cannot achieve multi-objective optimization and is inefficient, and is therefore not suitable for the feasibility analysis and preliminary design of energy efficiency improvement of wind-assisted propulsion ships, the present invention proposes a single sail layout optimization method for wind-assisted propulsion ships, which uses a method combining artificial neural networks and multi-objective genetic algorithms to obtain the optimal spatial layout, i.e., position, of a single sail in a given deck space.

[0008] The specific plan is as follows:

[0009] A single sail layout optimization method for a wind-assisted propulsion ship.

[0010] S1, parameter setting: given ship type, single sail, speed, and sea conditions as parameters:

[0011] S2, solving the performance evaluation model of wind-assisted propulsion ship: substituting the parameters in S1 into the performance evaluation model of wind-assisted propulsion ship to obtain the main engine power saving rate and the propulsion efficiency of the main propulsion device;

[0012] S3, solve the model by changing wind speed and wind direction: calculate the main engine power saving rate and the propulsion efficiency of the main propulsion device at different wind speeds and wind directions: use the global average route wind probability matrix provided by IMO to calculate the wind speed-wind direction average main engine power saving rate and the propulsion efficiency of the main propulsion device;

[0013] S4, solve the model by changing the position of a single sail: change the position of a single sail in a given deck space, and calculate the average main engine power saving rate and propulsion efficiency of the main propulsion device at different wind speed-wind direction positions of a single sail;

[0014] S5, training model based on artificial neural network algorithm: using artificial neural network algorithm, training a first proxy surface model of wind speed-wind direction average main engine power saving rate changing with the position of a single sail, and a second proxy surface model of wind speed-wind direction average main propulsion efficiency changing with the position of a single sail;

[0015] S6, calculating the optimal solution for multi-objective optimization: using a multi-objective genetic algorithm, taking the proxy surface model as the fitness of two objectives, calculating the Pareto optimal solution space of a single sail position as the optimal solution for multi-objective optimization.

[0016] Preferably, in step S2, the wind-assisted propulsion ship performance evaluation model includes: a four-degree-of-freedom mechanical equilibrium equation group constructed based on the set sea conditions and the wind-assisted propulsion ship coordinate system, as well as the relationship between the hydrodynamic and aerodynamic forces acting on the ship and the navigation attitude and effective thrust in the equation group, and the relationship between the relative rotation efficiency of the main propulsion device, the hull efficiency, the open water efficiency and the effective thrust and the propulsion efficiency, effective power, main engine power and main engine power saving rate.

[0017] Preferably, the four-degree-of-freedom mechanical equilibrium equations are:

[0018]

[0019]

[0020] In the formula, β represents the drift angle; δ represents the rudder angle; represents the heel angle; GM represents high static stability; Δ represents displacement; g represents gravitational acceleration; X CW represents hydrostatic resistance; X AW Indicates wave resistance increase; X Hand Y H They represent the additional resistance at drift angle and the lateral force respectively; CLR represents the distance from the point of action of the additional resistance at drift angle and the lateral force to the bow; H Indicates the Z coordinate of the point where the drift angle additional resistance and lateral force act; X W and Y W Respectively represent the windward forces on the hull and superstructure above the waterline; x W ,y W and z W They represent the x-, y-, and z-axis coordinates of the points on the hull and superstructure above the waterline where the windward force acts; S and Y S They represent the thrust and lateral force generated by the wind-assisted propulsion device respectively; x S ,y S and z S The x, y, and z axis coordinates of the points where the thrust and lateral force generated by the wind-assisted propulsion device are applied; R and Y R Respectively represent the rudder force in the x-axis and y-axis directions; R Indicates the Z coordinate of the point where the rudder force acts; T E Indicates the effective thrust of the main propulsion unit.

[0021] Preferably, the propulsion efficiency η of the main propulsion device is D The calculation method is:

[0022] η D =η R η H η O (5)

[0023] In the main propulsion unit, η R Relative rotation efficiency, η H represents the hull efficiency, η O represents the open water efficiency and η D Indicates propulsion efficiency.

[0024] Preferably, the effective power P of the main propulsion device E The calculation method is:

[0025] P E =T E V S (6)

[0026] Based on the four-degree-of-freedom mechanical equilibrium equations, the effective thrust T is obtained. E , V S Indicates the speed.

[0027] Preferably, under different wind directions, the main engine power P of the main propulsion device is S_total The calculation method is:

[0028]

[0029] Where η D represents propulsion efficiency; η S Indicates transmission efficiency; P S Indicates the power required for navigation; P S_rotor Indicates the power required to drive the rotor.

[0030] Preferably, the main engine power saving rate calculation method of the main propulsion device is:

[0031]

[0032] P S_total,w / o Sail Indicates the main engine power required without wind-assisted propulsion.

[0033] Preferably, in step S3, the wind speed values ​​include: 5m / s, 10m / s and 15m / s; the wind direction values ​​are taken at intervals of 5° and range from 0° to 360°.

[0034] Preferably, in step S6, in the multi-objective genetic algorithm, a proxy surface model of how the wind speed-wind direction average main engine power saving rate changes with the position of a single sail, a proxy surface model of how the wind speed-wind direction average main propulsion device propulsion efficiency changes with the position of a single sail, and constraints are used as input, and the Pareto optimal solution space of a single sail position is used as output, and the constraints include: the radius of the base of a single sail, the deck space, and the deck space available for arrangement.

[0035] Beneficial effects:

[0036] The present invention proposes a method for optimizing the layout of a single sail of a wind-assisted propulsion ship, which uses a method combining an artificial neural network (Artificial Neural Network) and a multi-objective genetic algorithm (Multi-objective Generic Algorithm), that is, using an artificial neural network to construct a proxy surface model of the relationship between the saving rate of the main engine power and the layout of a single sail, and a proxy surface model of the relationship between the propulsion efficiency of the main propulsion device and the layout of a single sail; then using the proxy surface as the fitness, using a multi-objective genetic algorithm, multi-objective optimization is performed on the layout of a single sail to obtain the optimal spatial layout (i.e., position) of a single sail in a given deck space for a given ship, a given single sail, and a given speed. The present invention can achieve multi-objective optimization, that is, taking into account both the maximization of the saving rate of the main engine power and the maximization of the propulsion efficiency of the main propulsion device; at the same time, it has a high efficiency and can be used for feasibility analysis and preliminary design of energy efficiency improvement of wind-assisted propulsion ships. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flow chart of a method for optimizing the layout of a single sail of a wind-assisted propulsion ship.

[0038] Figure 2 Coordinate system effect diagram of wind-assisted propulsion ship in the embodiment.

[0039] Figure 3 Effect diagram of the multi-objective genetic algorithm restriction conditions in the embodiment. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0041] Embodiment 1:

[0042] like Figure 1 As shown, the present invention provides a method for optimizing the layout of a single sail of a wind-assisted propulsion ship:

[0043] S1, parameter setting: given ship type, single sail, speed, and sea conditions as parameters:

[0044] S2, solving the performance evaluation model of wind-assisted propulsion ship: substituting the parameters in S1 into the performance evaluation model of wind-assisted propulsion ship to obtain the main engine power saving rate and the propulsion efficiency of the main propulsion device;

[0045] S3, solve the model by changing wind speed and wind direction: calculate the main engine power saving rate and the propulsion efficiency of the main propulsion device at different wind speeds and wind directions: use the global average route wind probability matrix provided by IMO to calculate the wind speed-wind direction average main engine power saving rate and the propulsion efficiency of the main propulsion device;

[0046] S4, solve the model by changing the position of a single sail: change the position of a single sail in a given deck space, and calculate the average main engine power saving rate and propulsion efficiency of the main propulsion device at different wind speed-wind direction positions of a single sail;

[0047] S5, training model based on artificial neural network algorithm: using artificial neural network algorithm, training a first proxy surface model of wind speed-wind direction average main engine power saving rate changing with the position of a single sail, and a second proxy surface model of wind speed-wind direction average main propulsion efficiency changing with the position of a single sail;

[0048] S6, calculating the optimal solution for multi-objective optimization: using a multi-objective genetic algorithm, taking the proxy surface model as the fitness of two objectives, calculating the Pareto optimal solution space of a single sail position as the optimal solution for multi-objective optimization.

[0049] Embodiment 2:

[0050] Given a ship type, a single sail, speed (Vs), wind (TWS and TWA) and waves (TWA, Hs and Tp):

[0051] (I) Solve the performance evaluation model of wind-assisted propulsion ships to obtain the main engine power saving rate Rsavings and the main propulsion device propulsion efficiency η D .

[0052] The performance evaluation model of wind-assisted propulsion ships includes a set of four-degree-of-freedom mechanical equilibrium equations, as well as the relationship between the hydrodynamic / aerodynamic forces on the ship and the navigation attitude (rudder angle, drift angle, heel angle) and effective thrust in the equations. The model is as follows.

[0053] The coordinate system of the wind-assisted propulsion ship is as follows Figure 2 As shown, the origin of the coordinate system is at the bow, the height is at the keel, the X-axis points to the stern, the Y-axis points to the starboard, and the Z-axis points vertically upward.

[0054] The positive and negative definitions of the ship (including appendages) motion and the wind direction angle are as follows:

[0055] (1) Drift angle - the angle between heading and heading, with counterclockwise being positive;

[0056] (2) Rudder angle: right rudder is positive;

[0057] (3) Heel angle: positive for left heel;

[0058] (4) Wind direction angle - the angle between the positive direction of the X-axis and the wind direction, with counterclockwise being positive.

[0059] According to this coordinate system and the definition of positive and negative, in a specific sea condition (wind: wind direction TWA, wind speed TWS; wave: three-in-one average wave height H 1 / 3 , peak period T p ) and specific speed V S Under this condition, the ship's four-degree-of-freedom mechanical equilibrium equations are as follows:

[0060]

[0061]

[0062] In the formula, β represents the drift angle; δ represents the rudder angle; represents the heel angle; GM represents high static stability; Δ represents displacement; g represents gravitational acceleration; X CW represents hydrostatic resistance; X AW Indicates wave resistance increase; X H and Y H They represent the additional resistance at drift angle and the lateral force respectively; CLR represents the distance from the point of action of the additional resistance at drift angle and the lateral force to the bow; H Indicates the Z coordinate of the point where the drift angle additional resistance and lateral force act; X W and Y W Respectively represent the windward forces on the hull and superstructure above the waterline; xW ,y W and z W They represent the x-, y-, and z-axis coordinates of the points on the hull and superstructure above the waterline where the windward force acts; S and Y S They represent the thrust and lateral force generated by the wind-assisted propulsion device respectively; x S ,y S and z S The x, y, and z axis coordinates of the points where the thrust and lateral force generated by the wind-assisted propulsion device are applied; R and Y R Respectively represent the rudder force in the x-axis and y-axis directions; R Indicates the Z coordinate of the point where the rudder force acts; T E Indicates the effective thrust of the main propulsion unit.

[0063] In this set of mechanical equilibrium equations, the relationship between hydrodynamic force, aerodynamic force, navigation attitude and effective thrust can be obtained by using empirical / semi-empirical formulas, or by three-dimensional CFD calculations or model tests.

[0064] This mechanical equilibrium equation group consists of four nonlinear equations. Solving β, δ, and T E The four unknown quantities constitute a mathematical closed system.

[0065] Use the Secant Method to solve β, δ, and T E Four unknown quantities. At the same time, the relative rotation efficiency η of the main propulsion device is solved by using the empirical formula R 、Hull efficiency η H and open water efficiency η O , and then use formula (5) to obtain the propulsion efficiency η of the main propulsion device D .

[0066] η D =η R η H η O (5)

[0067] Get effective thrust T E , and then obtain the effective power P E As shown in formula (6).

[0068] P E =T E V S (6)

[0069] According to formula (7), the required main engine power P under different wind directions is obtained: S_total .

[0070]

[0071] In the formula,

[0072] η D - Propulsion efficiency;

[0073] η S -Transmission efficiency;

[0074] P S - Power required for navigation;

[0075] P S_rotor -The power required to drive the rotor.

[0076] At the same time, calculate the required main engine power P without wind-assisted propulsion device S_total,w / o Sail , and then the host power saving rate is obtained, as shown in formula (8).

[0077]

[0078] (II) Change wind speed TWS and TWA, calculate the main engine power saving rate Rsavings and main propulsion efficiency η D .

[0079] The calculation examples to be calculated are shown in Table 1.

[0080] Table 1

[0081] Wind speed TWS (m / s) Wind direction TWA(°) 5.0 0.0-360.0, 5.0 intervals 10.0 0.0-360.0, 5.0 intervals 15.0 0.0-360.0, 5.0 intervals

[0082] (III) Using the global average route wind probability matrix provided by IMO:

[0083] IMO is derived from (IMO-MEPC.1-Circ.896-2021Guidance On Treatment Of InnovativeEnergy Efficiency Technologies for Calculation and Verification of theAttained EEDI and EEXI), which calculates the wind speed-wind direction average (TWS-and TWA-averaged) main engine power saving rate Rsavings and the main propulsion efficiency η D .

[0084] First, calculate the average (TWA-averaged) main engine power saving rate Rsavings and main propulsion efficiency η under different wind speeds TWS. D , as shown in equations (9)-(14).

[0085]

[0086] TWA-averaged Rsavings 15m / s = (11)

[0087] wa0Rsavings 15m / s,0 +wa5Rsavings 15m / s,5 +wa 10 Rsavings 15m / s,10 +...+wa 355

[0088] Rsavings 15m / s,355

[0089]

[0090] Among them, the weight wa is:

[0091] Table 2

[0092]

[0093]

[0094]

[0095] Then calculate the TWS-and TWA-averaged main engine power saving rate Rsavings and the main propulsion efficiency η D , as shown in formulas (15)-(16).

[0096]

[0097]

[0098] Among them, the weight ws value is shown in Table 3:

[0099] Table 3

[0100] <![CDATA[ws 5m / s ]]> 0.539 <![CDATA[ws 10m / s ]]> 0.3878 <![CDATA[ws 15m / s ]]> 0.0692

[0101] (iv) Change the position of a single sail in the available deck space (50 positions are recommended) and calculate the main engine power saving rate Rsavings and main propulsion efficiency η under different positions, the wind speed-wind direction average (TWS-and TWA-averaged) D .

[0102] (V) Using artificial neural network algorithm, two proxy surface models are trained: wind speed-wind direction average main engine power saving rate changes with the position of a single sail, and wind speed-wind direction average main propulsion device propulsion efficiency changes with the position of a single sail.

[0103] Divide the data collected in (IV) into a training set and a test set. If there are 50 sets of data in total, it is recommended to use 42 sets as training sets and the rest as test sets.

[0104] In the process of programming, the execution of artificial neural network algorithm can call the existing modules of programming tools. Taking Matlab as an example, the newff function can be called to construct the artificial neural network structure and return the artificial neural network type object. Calling the train function of the object can train the proxy face model with the training data set, calculate the accuracy of the proxy face model with the test set, and obtain R 2 , MAE and MSE three indicators.

[0105] (VI) Using a multi-objective genetic algorithm, with the above two proxy surface models as the fitness of the two objectives, the Pareto optimal solution space (Pareto Solutions) of a single sail position is calculated to achieve multi-objective optimization of the wind speed-wind direction average main engine power saving rate and the propulsion efficiency of the main propulsion device.

[0106] In the process of writing programs, the execution of multi-objective genetic algorithms can call existing modules of programming tools. Taking Matlab as an example, the gamultiobj function can be called, with the proxy surface model of wind speed-wind direction average main engine power saving rate changing with single sail position, the proxy surface model of wind speed-wind direction average main propulsion device propulsion efficiency changing with single sail position, and constraints as input, and the Pareto optimal solution space of single sail position as output. Figure 3 As shown, the multi-objective genetic algorithm constraints include: single sail base radius, deck space, and deck space available for layout.

[0107] It should be noted that the above-described specific implementation methods can enable those skilled in the art to more fully understand the present invention.

[0108] Therefore, although the present invention has been described in detail with reference to the drawings and embodiments, those skilled in the art should understand that the present invention can still be modified or replaced by equivalents. In short, all technical solutions and improvements that do not deviate from the spirit and scope of the present invention should be covered by the protection scope of the patent for the present invention.

Claims

1. A method for optimizing the layout of a single sail of a wind-assisted propulsion ship, characterized in that: S1, parameter setting: given ship type, single sail, speed, and sea conditions as parameters: S2, solving the performance evaluation model of wind-assisted propulsion ship: substituting the parameters in S1 into the performance evaluation model of wind-assisted propulsion ship to obtain the main engine power saving rate and the propulsion efficiency of the main propulsion device; S3, solve the model by changing wind speed and wind direction: calculate the main engine power saving rate and the propulsion efficiency of the main propulsion device at different wind speeds and wind directions: use the global average route wind probability matrix provided by IMO to calculate the wind speed-wind direction average main engine power saving rate and the propulsion efficiency of the main propulsion device; S4, solve the model by changing the position of a single sail: change the position of a single sail in a given deck space, and calculate the average main engine power saving rate and propulsion efficiency of the main propulsion device at different wind speed-wind direction positions of a single sail; S5, training model based on artificial neural network algorithm: using artificial neural network algorithm, training a first proxy surface model of wind speed-wind direction average main engine power saving rate changing with the position of a single sail, and a second proxy surface model of wind speed-wind direction average main propulsion efficiency changing with the position of a single sail; S6, calculating the optimal solution for multi-objective optimization: using a multi-objective genetic algorithm, taking the proxy surface model as the fitness of two objectives, calculating the Pareto optimal solution space of a single sail position as the optimal solution for multi-objective optimization.

2. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 1, characterized in that: In step S2, the wind-assisted propulsion ship performance evaluation model includes: a four-degree-of-freedom mechanical equilibrium equation group constructed based on the set sea conditions and the wind-assisted propulsion ship coordinate system, as well as the relationship between the hydrodynamic and aerodynamic forces acting on the ship and the navigation attitude and effective thrust in the equation group, and the relationship between the relative rotation efficiency, hull efficiency, open water efficiency and effective thrust of the main propulsion device and the propulsion efficiency, effective power, main engine power and main engine power saving rate.

3. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 2, characterized in that: The four-degree-of-freedom mechanical equilibrium equations are: Where, β represents the drift angle; δ represents the rudder angle; represents the heel angle; GM represents high static stability; Δ represents displacement; g represents gravitational acceleration; X CW represents hydrostatic resistance; X AW Indicates wave resistance increase; X H and Y H They represent the additional resistance at drift angle and the lateral force respectively; CLR represents the distance from the point of action of the additional resistance at drift angle and the lateral force to the bow; H Indicates the Z coordinate of the point where the drift angle additional resistance and lateral force act; X W and Y W Respectively represent the windward forces on the hull and superstructure above the waterline; x W ,y W and z W They represent the x-, y-, and z-axis coordinates of the points on the hull and superstructure above the waterline where the windward force acts; S and Y S They represent the thrust and lateral force generated by the wind-assisted propulsion device respectively; x S ,y S and z S The x, y, and z axis coordinates of the points where the thrust and lateral force generated by the wind-assisted propulsion device are applied; R and Y R Respectively represent the rudder force in the x-axis and y-axis directions; R Indicates the Z coordinate of the point where the rudder force acts; T E Indicates the effective thrust of the main propulsion unit.

4. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 2, characterized in that: The propulsion efficiency η of the main propulsion device D The calculation method is: or D =the R or H or O (5) In the main propulsion unit, η R Relative rotation efficiency, η H represents the hull efficiency, η O represents the open water efficiency and η D Indicates propulsion efficiency.

5. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 2, characterized in that: The effective power P of the main propulsion device E The calculation method is: P E =T E V S (6) Based on the four-degree-of-freedom mechanical equilibrium equations, the effective thrust T is obtained. E , V S Indicates the speed.

6. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 5, characterized in that: Under different wind directions, the main engine power P of the main propulsion device S_total The calculation method is: Where η D It indicates propulsion efficiency; η S Indicates transmission efficiency; P S Indicates the power required for navigation; P S_rotor Indicates the power required to drive the rotor.

7. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 6, characterized in that: The calculation method of the main engine power saving rate of the main propulsion device is: P S_total,w / oSail Indicates the main engine power required without wind-assisted propulsion.

8. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 1, characterized in that: In step S3, the wind speed values ​​include: 5m / s, 10m / s and 15m / s; the wind direction values ​​are taken at intervals of 5°, ranging from 0° to 360°.

9. A single sail layout optimization method for a wind-assisted propulsion ship according to claim 1, characterized in that: In step S6, in the multi-objective genetic algorithm, a proxy surface model of wind speed-wind direction average main engine power saving rate changing with a single sail position, a proxy surface model of wind speed-wind direction average main propulsion device propulsion efficiency changing with a single sail position and constraints are used as input, and the Pareto optimal solution space of a single sail position is used as output, wherein the constraints include: a single sail base radius, deck space, and deck space available for arrangement.

Images