Ship control motion hydrodynamic parameter identification method based on separated model

Abstract

The invention discloses a ship control motion hydrodynamic parameter identification method based on a separated model, and the method comprises the following steps: S1) obtaining control motion hydrodynamic parameters of a ship separated model based on ship navigation sample data; s2) preprocessing the sample data; s3) establishing an identification mathematical model; s4) processing the preprocessed sample data through a difference method and data centralization, redefining the input and output of a regression model for identifying the hydrodynamic coefficient, and obtaining re-identified control motion hydrodynamic parameters; and S5) applying the identified hydrodynamic parameters of the ship control motion to hydrodynamic calculation of a ship separation type model, and performing ship motion prediction or ship motion control based on the model. According to the method, a regression model for identifying the hydrodynamic coefficient is established for a sample, the unmanned ship hydrodynamic coefficient of a separated model is calculated, the regression model is reconstructed by adopting a difference method and a data center method so as to slow down parameter drift, and the hydrodynamic coefficient closer to a real value is obtained.

Description

Technical Field

[0001] The invention relates to ship motion control technology, and in particular to a method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model. Background Art

[0002] Ship maneuverability hydrodynamic parameters (hydrodynamic derivatives) are important parameters for ship motion prediction and are of great significance in the fields of ship design, ship maneuvering, ship motion control, etc. The main methods for obtaining ship hydrodynamic derivatives are: database or empirical formula method, constrained model test method, calculation method based on computational fluid dynamics (CFD) numerical simulation, and model test combined with system identification method, but the economic and time costs are relatively high.

[0003] Parameter identification methods are gradually being applied to ship maneuvering motion modeling, but they are basically used to identify the hydrodynamic derivatives of the overall ship model. Since the overall model is obtained by discretization near the design speed of the ship, it cannot be applied to the situation where the speed deviates from the design speed, so it cannot be effectively applied in the fields of ship speed control. In addition, the commonly used identification methods currently include least squares method, extended Kalman filtering method, maximum likelihood estimation, etc. For small sample problems, the above identification methods often lead to overfitting, resulting in poor generalization ability and low accuracy of the identified mathematical model, which makes it difficult to accurately predict the ship's motion. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model in view of the defects in the prior art.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model, comprising the following steps:

[0006] S1) obtaining the hydrodynamic parameters of the maneuvering motion of the ship separation model based on the ship navigation sample data;

[0007] S2) preprocessing the sample data;

[0008] The preprocessing of sample data includes two steps: purification and cleaning, which include interval extraction and threshold filtering;

[0009] S3) establishing a mathematical model for identification;

[0010] A regression model for hydrodynamic coefficient identification is established for the samples, and the hydrodynamic parameters of the unmanned boat of the separation model are calculated;

[0011] The motion state variable vector is used as input, the hydrodynamic vector is used as output, and the motion hydrodynamic parameter vector is used as the vector to be identified;

[0012] S4) processing the preprocessed sample data by difference method and data centering, redefining the input and output of the regression model for hydrodynamic coefficient identification, and obtaining the re-identified hydrodynamic parameters of the maneuvering motion;

[0013] Slow down the linear relationship between the hydrodynamic derivative vectors and reduce parameter drift;

[0014] S5) applying the identified ship maneuvering hydrodynamic coefficients to the hydrodynamic calculation of the ship separation model, which can be used for ship motion prediction or model-based ship motion control.

[0015] According to the above scheme, in step S1), the three-degree-of-freedom separation model of the ship horizontal plane is:

[0016]

[0017] Among them, u, v, and r are the forward speed, rolling speed, and rolling angular velocity of the ship respectively; is the heading angle, δ is the spray angle, X, Y and N represent the external forces and moments of the ship in three directions respectively, the subscript P represents the force and thrust moment generated by the water jet propulsion, and the subscript H represents the viscous force and moment of the hull; x and y are the positions of the ship in the earth-fixed coordinate system, m is the hull mass, m x is the longitudinal additional mass, m y is the lateral additional mass, I zz is the yaw moment of inertia; J zz is the additional moment of inertia in yaw;

[0018] They are x, y, Derivatives of u, v, r;

[0019] According to the above scheme, in step S3), the identification mathematical model established is:

[0020]

[0021] In the formula, the hydrodynamic coefficient vector (X w ,Y w ,N w ) and the motion state variable vector (C Xw ,C Yw ,C Nw ) are:

[0022] X w =[Xuu′,Xvv′,Xrr′] T

[0023] Y w =[Yv′,Yr′,Yvv′,Yvr′,Yrr′] T

[0024] N w =[Nv′,Nr′,Nrr′,Nvvr′,Nvrr′] T

[0025] C Xw =[u 2 (k),v 2 (k),r 2 (k)L 2 ]

[0026]

[0027] According to the above scheme, in step S4), the pre-processed sample data is processed by difference method and data centering, and the input and output of the regression model for hydrodynamic coefficient identification are redefined as follows:

[0028] S4.1) performing differential processing on the preprocessed sample data;

[0029] The difference method is:

[0030] y(k)=a1x1(k)+a2x2(k)+…+a n x n (k);

[0031] Where k is the sample sampling time, y(k) is the output, and x(k) is the input, x(k) = (x1(k), x2(k), ..., x n (k));

[0032] Then at time k-1:

[0033] y(k-1)=a1x1(k-1)+a2x2(k-1)+...+a n x n (k-1)

[0034] Subtract the above equation and multiply it by the constant ρ to get:

[0035] y(k)-ρy(k-1)=a1[x1(k)-ρx1(k-1)]+a2[x2(k)-ρx2(k-1)]+...+a n [x n (k)-ρx n (k-1)]

[0036] Redefine y(k)-ρy(k-1) as output and x(k)-ρx(k-1) as input.

[0037] Apply the difference method to redefine the input and output of the regression model;

[0038] The hydrodynamic coefficient vector (X w ,Y w ,N w ) remains unchanged, while the motion state variable vector (C XW ,C YW ,C NW ) is corrected to [C Xw * ,C Yw * ,C Nw * ]:

[0039] C Xw * =[u 2 (k)-u 2 (k-1),v 2 (k)-v 2 (k-1),r 2 (k)L 2 -r 2 (k-1)L 2 ];

[0040] C Yw * =[v(k)U(k)-v(k-1)U(k-1),r(k)U(k)Lr(k-1)U(k-1)L,v(k)|v(k) |-v(k-1)|v(k-1)|,|v(k)|r(k)L-|v(k-1)|r(k-1)L,r(k)|r(k)|L 2 -r(k-1)|r(k-1)|L 2 ]

[0041] C Nw * =[v(k)U(k)-v(k-1)U(k-1),r(k)U(k)Lr(k-1)U(k-1)L,r(k)|r(k)|L 2 -r(k-1)|r(k-1)|L 2 ,v 2 (k)r(k)L / U(k)-v 2 (k-1)r(k-1)L / U(k-1),v(k)r 2 (k)L 2 U(k)-v(k-1)r 2 (k-1)L2 / U(k-1)]

[0042] The output hydrodynamic vector of the sample after using the difference method is as follows:

[0043]

[0044] 4.2) Data centering is: using the sample data of the variable minus the mean of the data, that is:

[0045]

[0046] in, is the centralized data, x i is the original data, μ is the mean of the sample data;

[0047] After data centering, the input and output of the regression model for hydrodynamic coefficient identification are redefined as follows:

[0048] Input: [C Xw * ,C Yw * ,C Nw * ];

[0049] Output:

[0050]

[0051] The beneficial effects produced by the present invention are:

[0052] The present invention establishes a regression model for hydrodynamic coefficient identification for samples, calculates the hydrodynamic coefficient of the unmanned boat of the separation model, and reconstructs the regression model using the difference method and the data center method to slow down the parameter drift phenomenon and obtain a hydrodynamic coefficient that is closer to the true value. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0054] Figure 1 It is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0056] like Figure 1As shown, a method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model comprises the following steps:

[0057] S1) obtaining the hydrodynamic parameters of the maneuvering motion of the ship separation model based on the ship navigation sample data;

[0058] The data samples were obtained through simulation tests, and the three-degree-of-freedom separation model of the ship's horizontal plane is:

[0059]

[0060] Among them, u, v, and r are the forward speed, rolling speed, and rolling angular velocity of the ship respectively; is the heading angle, δ is the spray angle, X, Y and N represent the external forces and moments of the ship in three directions respectively, the subscript P represents the force and thrust moment generated by the water jet propulsion, and the subscript H represents the viscous force and moment of the hull; x and y are the positions of the ship in the earth-fixed coordinate system, m is the hull mass, m x is the longitudinal additional mass, m y is the lateral additional mass, I zz is the yaw moment of inertia; J zz is the additional moment of inertia in yaw;

[0061] They are x, y, Derivatives of u, v, r;

[0062] S2) preprocessing the sample data;

[0063] The preprocessing of sample data includes two steps: purification and cleaning, which include interval extraction and threshold filtering;

[0064] S3) establishing a mathematical model for identification;

[0065] A regression model for hydrodynamic coefficient identification is established for the samples, and the hydrodynamic parameters of the unmanned boat of the separation model are calculated;

[0066] The motion state variable vector is used as input, the hydrodynamic vector is used as output, and the motion hydrodynamic parameter vector is used as the vector to be identified;

[0067] The established identification mathematical model is:

[0068]

[0069] In the formula, the hydrodynamic coefficient vector (X w ,Y w ,N w ) and the motion state variable vector (C Xw ,C Yw ,C Nw ) are:

[0070] X w =[Xuu′,Xvv′,Xrr′] T

[0071] Y w =[Yv′,Yr′,Yvv′,Yvr′,Yrr′] T

[0072] N w =[Nv′,Nr′,Nrr′,Nvvr′,Nvrr′] T

[0073] C Xw =[u 2 (k),v 2 (k),r 2 (k)L 2 ]

[0074]

[0075] Among them, m' is the hull mass, m x ' is the longitudinal additional mass, m y ' is the lateral additional mass;

[0076] S4) processing the preprocessed sample data by difference method and data centering, redefining the input and output of the regression model for hydrodynamic coefficient identification, and obtaining the re-identified hydrodynamic parameters of the maneuvering motion;

[0077] Slow down the linear relationship between the hydrodynamic derivative vectors and reduce parameter drift;

[0078] The details are as follows:

[0079] S4.1) performing differential processing on the preprocessed sample data;

[0080] The difference method is:

[0081] y(k)=a1x1(k)+a2x2(k)+...+a n x n (k);

[0082] Where k is the sample sampling time, y(k) is the output, and x(k) is the input, x(k) = (x1(k), x2(k), ..., x n (k));

[0083] Then at time k-1:

[0084] y(k-1)=a1x1(k-1)+a2x2(k-1)+...+a nx n (k-1)

[0085] Subtract the above equation and multiply it by the constant ρ to get:

[0086] y(k)-ρy(k-1)=a1[x1(k)-ρx1(k-1)]+a2[x2(k)-ρx2(k-1)]+…+a n [x n (k)-ρx n (k-1)]

[0087] Redefine y(k)-y(k-1) as output and x(k)-ρx(k-1) as input.

[0088] Apply the difference method to redefine the input and output of the regression model;

[0089] The hydrodynamic coefficient vector (X w ,Y w ,N w ) remains unchanged, while the motion state variable vector (C XW ,C YW ,C NW ) is corrected to [C Xw * ,C Yw * ,C Nw * ]:

[0090] C Xw * =[u 2 (k)-u 2 (k-1),v 2 (k)-v 2 (k-1),r 2 (k)L 2 -r 2 (k-1)L 2 ];

[0091] C Yw * =[v(k)U(k)-v(k-1)U(k-1),r(k)U(k)Lr(k-1)U(k-1)L,v(k)|v(k) |-v(k-1)|v(k-1)|,|v(k)|r(k)L-|v(k-1)|r(k-1)L,r(k)|r(k)|L 2 -r(k-1)|r(k-1)|L 2 ]

[0092] C Nw *=[v(k)U(k)-v(k-1)U(k-1),r(k)U(k)Lr(k-1)U(k-1)L,r(k)|r(k)|L 2 -r(k-1)|r(k-1)|L 2 ,v 2 (k)r(k)L / U(k)-v 2 (k-1)r(k-1)L / U(k-1),v(k)r 2 (k)L 2 U(k)-v(k-1)r 2 (k-1)L 2 / U(k-1)] The output hydrodynamic vector of the sample after using the difference method is as follows:

[0093]

[0094] 4.2) Data centering is: subtract the mean of the data from the sample data of a variable, that is:

[0095]

[0096] in, is the centralized data, x i is the original data, μ is the mean of the sample data;

[0097] After data centering, the input and output of the regression model for hydrodynamic coefficient identification are redefined as follows:

[0098] Input: [C Xw * ,C Yw * ,C Nw * ];

[0099] Output:

[0100]

[0101] S5) applying the identified motion hydrodynamic parameters to the hydrodynamic calculation of the ship separation model, which can be used for ship motion prediction or model-based ship motion control.

[0102] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model, characterized in that: The following steps are involved: S1) obtaining the hydrodynamic parameters of the maneuvering motion of the ship separation model based on the ship navigation sample data; S2) preprocessing the sample data; S3) establishing a mathematical model for identification; A regression model for hydrodynamic coefficient identification is established for the samples, and the hydrodynamic parameters of the unmanned boat of the separation model are calculated; The motion state variable vector is used as input, the hydrodynamic vector is used as output, and the hydrodynamic parameters of the manipulation motion are used as the vector to be identified; S4) processing the preprocessed sample data by difference method and data centering, redefining the input and output of the regression model for hydrodynamic coefficient identification, and obtaining the re-identified hydrodynamic parameters of the maneuvering motion; S5) applying the identified hydrodynamic parameters of the ship maneuvering motion to the hydrodynamic calculation of the ship separation model to perform ship motion prediction or model-based ship motion control.

2. According to the method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model according to claim 1, in the step S1), the three-degree-of-freedom separation model of the ship horizontal plane is: in, u, v, and r are the ship’s forward speed, rolling speed, and rolling angular velocity, respectively; is the heading angle, δ is the spray angle, X, Y and N represent the external forces and moments of the ship in three directions respectively, the subscript P represents the force and thrust moment generated by the water jet propulsion, and the subscript H represents the viscous force and moment of the hull; x and y are the positions of the ship in the earth-fixed coordinate system, m is the hull mass, m x is the longitudinal additional mass, m y is the lateral additional mass, I zz is the yaw moment of inertia; J zz is the additional moment of inertia in yaw; They are x, y, Derivatives of u, v, r.

3. The method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model according to claim 1, characterized in that: In the step S2), the preprocessing of the sample data includes two steps of purification and cleaning, and the methods include interval extraction and threshold filtering.

4. The method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model according to claim 2, characterized in that: In the step S3), the identification mathematical model established is: In the formula, the hydrodynamic coefficient vector (X w ,Y w ,N w ) and the motion state variable vector (C Xw ,C Yw ,C Nw ) are: X w =[Xuu′,Xvv′,Xrr′] T Y w =[Yv′,Yr′,Yvv′,Yvr′,Yrr′] T N w =[Nv′,Nr′,Nrr′,Nvvr′,Nvrr′] T C Xw =[u 2 (k),v 2 (k),r 2 (k)L 2 ] C Yw =[v(k)U(k),r(k)U(k)L,v(k)|v(k)|,|v(k)|r(k)L,r(k)|r(k)|L 2 ] 5. The method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model according to claim 1 or 4, characterized in that: In the step S4), the pre-processed sample data is processed by difference method and data centering, and the input and output of the regression model for hydrodynamic coefficient identification are redefined as follows: S4.1) performing differential processing on the preprocessed sample data; The difference method is: y(k)=a1x1(k)+a2x2(k)+…+a n X n (k); Where k is the sample sampling time, x(k) is the input to be processed, x(k) = x1(k), x2(k), ..., x n (k); y(k) is the processed output; Then at time k-1: y(k-1)=a1x1(k-1)+a2x2(k-1)+...+a n X n (k-1) Subtract the above equation and multiply it by the constant ρ to get: y(k)-ρy(k-1)=a1[x1(k)-ρx1(k-1)]+a2[x2(k)-ρx2(k-1)] +...+a n [x n (k)-ρx n (k-1)] Redefine y(k)-ρy(k-1) as output and x(k)-ρx(k-1) as input; Then the above difference method is applied to redefine the input and output of the regression model; S4.2) Data centering refers to the sample data of a variable minus the mean of the data, that is: in, is the centralized data, x i is the original data, μ is the mean of the sample data; After data centering, the input and output of the regression model for hydrodynamic coefficient identification are redefined by subtracting the mean of the vector data from the sample vector data.

6. The method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model according to claim 5, characterized in that: Applying the difference method, the input and output of the regression model are redefined as; The hydrodynamic coefficient vector (X w ,Y w ,N w ) remains unchanged, while the motion state variable vector (C XW ,C YW ,C NW ) is corrected to [C Xw * ,C Yw * ,C Nw * ]: C Xw * =[u 2 (k)-u 2 (k-1),v 2 (k)-v 2 (k-1),r 2 (k)L 2 -r 2 (k-1)L 2 ]; C Yw * =[v(k)U(k)-v(k-1)U(k-1),r(k)U(k)L-r(k-1)U(k-1)L,v(k)|v(k)|-v(k-1)|v(k-1)|, |v(k)|r(k)L-|v(k-1)|r(k-1)L,r(k)|r(k)|L 2 -r(k-1)|r(k-1)|L 2 ] C Nw * =[v(k)U(k)-v(k-1)U(k-1),r(k)U(k)L-r(k-1)U(k-1)L,r(k)|r(k)|L 2 -r(k-1)|r(k-1)|L 2 , v 2 (k)r(k)L / U(k)-v 2 (k-1)r(k-1)L / U(k-1),v(k)r 2 (k)L 2 U(k)-v(k-1)r 2 (k-1)L 2 / U(k-1)] The output hydrodynamic vector of the sample after using the difference method is as follows:

7. The method for identifying hydrodynamic parameters of ship maneuvering motion based on a separation model according to claim 5, characterized in that: The input and output of the regression model for hydrodynamic coefficient identification are redefined as: Input: [C Xw * ,C Yw * ,C Nw * ]; Output:

8. An electronic device, characterized in that: include: one or more processors; as well as a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images