CN110794682A - Thrust distribution method for multi-propeller rotatable ship - Google Patents

Abstract

The invention provides a thrust distribution method for a rotatable multi-propeller ship, which comprises the steps of firstly establishing a ship three-degree-of-freedom thrust distribution model, then establishing a multi-objective optimization objective function and constraint conditions related to propeller energy consumption, mechanical wear, singularity and thrust error, and finally solving the provided problem by utilizing an improved fast non-dominated genetic algorithm with an elite reservation strategy. The advantages are that: the energy consumption and the mechanical loss of the propulsion system can be effectively reduced, and the precision of ship control can be ensured.

Description

Thrust distribution method for multi-propeller rotatable ship

Technical Field

Relates to the field of ship control, in particular to a thrust distribution method for a rotatable multi-propeller ship.

Background

The rotatable propeller can rotate around an axis, can obtain maximum thrust in any direction, and can enable the ship to rotate in situ, move transversely, retreat rapidly and other special driving operations. The rotary propeller is suitable for various engineering ships, such as tugboats, floating crane ships, dredge ships, ferries, flat-bottomed boats for operation and the like, and has wide market application prospect and military significance. In order to improve the maneuverability, efficiency and system reliability of these engineering vessels, more than two propellers are typically provided. On one hand, however, due to the interaction of the ship navigation inertia force and the propeller thrust, the propulsion operation state of the multi-propeller presents asymmetric and irregular power disturbance characteristics, and the superposition of the irregular characteristics of the propellers also brings difficulty to the ship course control; on the other hand, the propulsion load fluctuates severely under the action of sea conditions, so that the power distribution among the propellers is greatly unbalanced, which brings difficulty to the speed control of the ship. Therefore, how to accurately distribute the thrust and angle of each rotatable propeller becomes a critical issue.

Disclosure of Invention

The invention provides a thrust distribution method for a rotary multi-propeller ship, which comprises the steps of firstly establishing a ship three-degree-of-freedom thrust distribution model according to conditions, then establishing a multi-objective optimization objective function and constraint conditions related to propeller energy consumption, abrasion, singularity and thrust error, and finally solving the provided problem by utilizing an improved rapid non-dominated multi-objective optimization algorithm with an elite reservation strategy.

The method mainly comprises the following steps:

step 1, establishing a three-degree-of-freedom thrust distribution model;

the motions in the heave, roll and pitch directions are ignored, and the ship dynamics model of the three-degree-of-freedom motions in the heave, roll and yaw directions can be summarized as follows:

wherein η ═ x, y, ψ]^TFor vessels in the geodetic coordinate system X_EOY_EActual position vector of, η_dIs the desired position vector. v ═ u, v, r]^TIs the actual velocity vector v under the hull coordinate system XOY_dIs the desired velocity vector. R (psi) is a rotation matrix, M is an inertia matrix of the hull, C (v) represents a Coriolis centripetal force matrix, and D (v) is a damping matrix. Tau is the actual resultant force generated by the ship propulsion systemAnd resultant moment, τ_dThe desired resultant force and resultant moment.

For a ship equipped with a rotatable multi-propeller, tau consists of the working state and thrust structure of the propeller:

τ＝B(α)f＝B(α)[f₁,f₂,f₃,f₄]^T(3)

wherein B (α) ∈ R^3×4A thrust structural matrix, f is a thrust matrix formed by the thrust of each propeller, f_iIs the thrust of propeller i. The thrust structure matrixes of the full-rotation propeller and the ducted propeller are respectively B_azi(α_i) And B_tun：

B(α)＝[B_azi(α₁),B_azi(α₂),B_azi(α₃),B_tun](5)

i is the number of propeller, i is 1,2,3,4, α_iIs the angle of the propeller i. l_i＝[l_xi,l_yi]^TFor the position coordinates of the propeller i mounted on the hull under the hull coordinate system XOY,/_xiIs the X coordinate of propeller i, l_yiIs the Y coordinate of propeller i.

Step 2, establishing a thrust distribution constraint condition;

wherein the maximum output forward thrust and the maximum output reverse thrust of the propeller i are respectively f_imaxAnd f_imin，f_i0Is the thrust generated by the propeller i of the preceding step, f_iα, the thrust at the present time_iminAnd α_imaxUpper and lower limits of propeller rotation angle, α respectively_i0Angle of propeller i of the previous step, α_iIs the angle of the propeller at the present moment. Δ f_iminAnd Δ f_imaxLower and upper limits, respectively, of the rate of change of thrust of propeller i. Δ α_iminAnd Δ α_imaxRespectively, the lower limit and the upper limit of the rate of change of the angle of the propeller i.

Step 3, establishing a thrust distribution optimization target;

wherein, P_wFor the overall power consumption of the multiple propellers, c_iIs the power coefficient of propeller i; j. the design is a square_eAnd the thrust error penalty term is obtained, s is the error between the actual thrust and the ideal thrust, and Q is the weight matrix of the error penalty term. J. the design is a square_ΩAnd omega is a weight matrix of the angle change penalty term. J. the design is a square_sThe matrix is a singular structure punishment item, delta is a weight matrix of the singular structure punishment item, and epsilon is a constant.

Step 4, solving a thrust distribution objective function by using an improved fast non-dominated sorting genetic algorithm with an elite retention strategy, and comprising the following substeps:

4-1) selecting the angle and the thrust of the propeller as decision variables, and designing a chromosome Ch ═ { f₁,f₂,f₃,f₄,α₁,α₂,α₃}; initializing population, generating parent population P with population size N_tAnd the initial value of t is 0;

4-2) Pair of parent population P_tPerforming polynomial mutation to generate progeny population Q_t；

4-3) the parent population P_tAnd the offspring population Q_tPerforming fusion to obtain a temporary fusion population R_t；

4-4) to R_tPerforming fast non-dominant sequencing to obtain different pareto frontiers F_j. N individuals in the population are divided into at most N pareto fronts, i.e., j ═ 1,2.. N;

4-5) for each leading edge F_jIn descending order of the crowding distance, crowding distance D_kThe calculation method of (c) is as follows:

wherein, | F_jL is the leading edge F_jNumber of individuals in (a).

4-6)P_t+1Has no individuals in the initial population of (1). Selection of F_jIs a first N-P_t+1Put in P_t+1In (1). If F_j+P_t+1＜N，P_t+1＝P_t+1∪F_jJ equals j +1, go back to execution 4-4); otherwise, returning to execute 4-5);

4-7)G_maxfor the maximum algebra of evolution, if t is greater than or equal to G_maxOutputting an optimal solution set, and finishing an optimization algorithm; otherwise, t is t +1, and P is added_tPerforming cross and differential variation operation to generate a population Q_tAnd step 4-3) is executed in a circulating way until the end. The differential variation mode is as follows:

Q_t＝βP_best+(1-β)P_t(9)

wherein β ∈ [0,1 ]]，P_bestIs a leading edge F₁The best individual in (1).

The method has the following effects and advantages:

the thrust obtained by the proposed thrust distribution method does not frequently jump, and the accurate tracking of the expected force and moment of the ship can be realized. The thrust distribution method controls the angle of the propeller not to be changed repeatedly, and avoids the mechanical abrasion of the propelling device; the energy consumption of the propulsion system can be effectively reduced, and the accuracy and the real-time performance of ship control can be ensured.

Drawings

FIG. 1 is a schematic view of thrust distribution for a multi-propeller vessel

FIG. 2 is a schematic view of a multi-propeller distribution

FIG. 3 is a flow chart of an improved fast non-dominated sorting genetic algorithm with elite retention strategy

Detailed Description

The invention provides a thrust distribution method for a rotary multi-propeller ship, which comprises the steps of firstly establishing a ship three-degree-of-freedom thrust distribution model according to conditions, then establishing a multi-objective optimization objective function and constraint conditions related to propeller energy consumption, abrasion, singularity and thrust error, and finally solving the provided problem by utilizing an improved rapid non-dominated multi-objective optimization algorithm with an elite reservation strategy. The method comprises the following steps:

step 1, establishing a three-degree-of-freedom thrust distribution model;

the motions in the heave, roll and pitch directions are ignored, and the ship dynamics model of the three-degree-of-freedom motions in the heave, roll and yaw directions can be summarized as follows:

wherein η ═ x, y, ψ]^TFor vessels in the geodetic coordinate system X_EOY_EActual position vector of, η_dIs the desired position vector. v ═ u, v, r]^TIs the actual velocity vector v under the hull coordinate system XOY_dIs the desired velocity vector. R (psi) is a rotation matrix, M is an inertia matrix of the hull, C (v) represents a Coriolis centripetal force matrix, and D (v) is a damping matrix. Tau is the actual resultant force and resultant moment generated by the ship propulsion system, tau_dThe desired resultant force and resultant moment.

As shown in fig. 1, the purpose of the thrust force distribution algorithm is to distribute the control force and moment calculated by the ship motion controller to each propeller so that the actual resultant force and resultant moment τ generated by each propeller can satisfy τ_d。

For a ship equipped with rotatable multiple propellers, the arrangement of the propellers is shown in fig. 2, and tau is composed of the working state and the thrust structure of the propellers:

τ＝B(α)f＝B(α)[f₁,f₂,f₃,f₄]^T(3)

wherein B (α) ∈ R^3×4A thrust structural matrix, f is a thrust matrix formed by the thrust of each propeller, f_iIs the thrust of propeller i. The ducted propeller is pushed by the fixed direction of the ducted propellerThe force structure matrix is only related to the position of the propeller. The thrust structure matrixes of the full-rotation propeller and the ducted propeller are respectively B_azi(α_i) And B_tun：

B(α)＝[B_azi(α₁),B_azi(α₂),B_azi(α₃),B_tun](5)

i is the number of propeller, i is 1,2,3,4, α_iIs the angle of the propeller i. l_i＝[l_xi,l_yi]^TFor the position coordinates of the propeller i mounted on the hull under the hull coordinate system XOY,/_xiIs the X coordinate of propeller i, l_yiIs the Y coordinate of propeller i.

Step 2, establishing a thrust distribution constraint condition;

wherein the maximum output forward thrust and the maximum output reverse thrust of the propeller i are respectively f_imaxAnd f_imin，f_i0Is the thrust generated by the propeller i of the preceding step, f_iα, the thrust at the present time_iminAnd α_imaxUpper and lower limits of propeller rotation angle, α respectively_i0Angle of propeller i of the previous step, α_iIs the angle of the propeller at the present moment. Δ f_iminAnd Δ f_imaxLower and upper limits, respectively, of the rate of change of thrust of propeller i, Δ α_iminAnd Δ α_imaxRespectively, the lower limit and the upper limit of the rate of change of the angle of the propeller i.

Step 3, establishing a thrust distribution optimization target;

wherein, P_wFor the overall power consumption of the multiple propellers, c_iBeing propellers iA power coefficient; j. the design is a square_eAnd the thrust error penalty term is obtained, s is the error between the actual thrust and the ideal thrust, and Q is the weight matrix of the error penalty term. J. the design is a square_ΩAnd omega is a weight matrix of the angle change penalty term. J. the design is a square_sThe matrix is a singular structure punishment item, delta is a weight matrix of the singular structure punishment item, and epsilon is a constant. The aim of the thrust distribution is to reduce the overall power consumption of the ship propeller and to avoid mechanical wear and singular thrust structures, while ensuring sufficiently small thrust errors.

And 4, solving a thrust distribution objective function by using an improved fast non-dominated sorting genetic algorithm with an elite retention strategy, wherein the flow chart of the algorithm is shown in FIG. 3. Step 4 comprises the following substeps:

4-1) selecting the angle and the thrust of the propeller as decision variables, and designing a chromosome Ch ═ { f₁,f₂,f₃,f₄,α₁,α₂,α₃}; initializing population, generating parent population P with population size N_tAnd the initial value of t is 0;

4-2) Pair of parent population P_tPerforming polynomial mutation to generate progeny population Q_t；

4-3) the parent population P_tAnd the offspring population Q_tPerforming fusion to obtain a temporary fusion population R_t；

4-4) to R_tPerforming fast non-dominant sequencing to obtain different pareto frontiers F_j. N individuals in the population are divided into at most N pareto fronts, i.e., j ═ 1,2.. N;

4-5) for each leading edge F_jIn descending order of the crowding distance, crowding distance D_kThe calculation method of (c) is as follows:

wherein, | F_jL is the leading edge F_jNumber of individuals in (a).

4-6)P_t+1Has no individuals in the initial population of (1). Selection of F_jIs a first N-P_t+1Put in P_t+1In (1). If F_j+P_t+1＜N，P_t+1＝P_t+1∪F_jJ equals j +1, go back to execution 4-4); otherwise, returning to execute 4-5);

4-7)G_maxfor the maximum algebra of evolution, if t is greater than or equal to G_maxOutputting an optimal solution set, and finishing an optimization algorithm; otherwise, t is t +1, and P is added_tPerforming cross and differential variation operation to generate a population Q_tAnd step 4-3) is executed in a circulating way until the end. The differential variation mode is as follows:

Q_t＝βP_best+(1-β)P_t(9)

wherein β ∈ [0,1 ]]，P_bestIs a leading edge F₁The best individual in (1).

Claims (1)

1. A thrust force distribution method for a rotatable multi-propeller ship, characterized by:

step 1, establishing a three-degree-of-freedom thrust distribution model;

the motions in the heave, roll and pitch directions are ignored, and the ship dynamics model of the three-degree-of-freedom motions in the heave, roll and yaw directions can be summarized as follows:

wherein η ═ x, y, ψ]^TFor vessels in the geodetic coordinate system X_EOY_EThe actual position vector of the lower one,

first derivative of η, η_dIs a desired position vector; v ═ u, v, r]^TIs the actual speed vector under the hull coordinate system XOY,

is the first derivative of v, v_dIs a desired velocity vector; r (psi) is a rotation matrix, M is an inertia matrix of the ship body, C (v) represents a Coriolis centripetal force matrix, and D (v) is a damping matrix; tau is the actual resultant force and resultant moment generated by the ship propulsion system, tau_dThe desired resultant force and resultant moment;

for a ship equipped with a rotatable multi-propeller, tau consists of the working state and thrust structure of the propeller:

τ＝B(α)f＝B(α)[f₁,f₂,f₃,f₄]^T(3)

wherein B (α) ∈ R^3×4A thrust structural matrix, f is a thrust matrix formed by the thrust of each propeller, f_iIs the thrust of propeller i; the thrust structure matrixes of the full-rotation propeller and the ducted propeller are respectively B_azi(α_i) And B_tun：

B(α)＝[B_azi(α₁),B_azi(α₂),B_azi(α₃),B_tun](5)

i is the number of propeller, i is 1,2,3,4, α_iIs the angle of the propeller i; l_i＝[l_xi,l_yi]^TFor the position coordinates of the propeller i mounted on the hull under the hull coordinate system XOY,/_xiIs the X coordinate of propeller i, l_yiIs the Y coordinate of propeller i;

step 2, establishing a thrust distribution constraint condition;

wherein the maximum output forward thrust and the maximum output reverse thrust of the propeller i are respectively f_imaxAnd f_imin，f_i0Is the thrust generated by the propeller i of the preceding step, f_iIs the thrust at the present moment α_iminAnd α_imaxAre respectively a screwUpper and lower limits of propeller rotation angle, α_i0Angle of propeller i of the previous step, α_iIs the angle of the propeller at the present moment; Δ f_iminAnd Δ f_imaxLower and upper limits, respectively, of the rate of change of thrust of propeller i, Δ α_iminAnd Δ α_imaxRespectively, the lower limit and the upper limit of the angle change rate of the propeller i;

step 3, establishing a thrust distribution optimization target;

wherein, P_wFor the overall power consumption of the multiple propellers, c_iIs the power coefficient of propeller i; j. the design is a square_eA thrust error punishment item is obtained, s is an error between actual thrust and ideal thrust, and Q is a weight matrix of the error punishment item; j. the design is a square_ΩThe weight matrix is an angle change penalty term, and omega is a weight matrix of the angle change penalty term; j. the design is a square_sThe matrix is a singular structure punishment item, delta is a weight matrix of the singular structure punishment item, and epsilon is a constant;

step 4, solving a thrust distribution objective function by using an improved fast non-dominated sorting genetic algorithm with an elite retention strategy, and comprising the following substeps:

4-1) selecting the angle and the thrust of each propeller as decision variables to design a chromosome; initializing population, generating parent population P with population size N_tAnd the initial value of t is 0;

4-2) Pair of parent population P_tPerforming polynomial mutation to generate progeny population Q_t；

4-3) the parent population P_tAnd the offspring population Q_tPerforming fusion to obtain a temporary fusion population R_t；

4-4) to R_tPerforming fast non-dominant sequencing to obtain different pareto frontiers F_jThe maximum value of j is N;

4-5) for each leading edge F_jThe individuals in (1) are arranged in descending order according to the crowding distance; crowding distance D_kThe calculation method of (c) is as follows:

wherein, | F_jL is the leading edge F_jThe number of individuals in (a);

4-6)P_t+1no individuals are present in the initial population of (a); selection of F_jIs a first N-P_t+1Put in P_t+1Performing the following steps; if F_j+P_t+1＜N，P_t+1＝P_t+1∪F_jI is i +1, go back to execution 4-4); otherwise, returning to execute 4-5);

4-7)G_maxfor the maximum algebra of evolution, if t is greater than or equal to G_maxOutputting an optimal solution set P_tThe optimization algorithm is ended; otherwise, t is t +1, and P is added_tPerforming cross and differential variation operation to generate a population Q_tAnd circularly executing the step 4-3) until the end; the differential variation mode is as follows:

Q_t＝βP_best+(1-β)P_t(9)

wherein β ∈ [0,1 ]]，P_bestIs a leading edge F₁The best individual in (1).

Images