CN113156815B - Data-driven marine ship motion attitude real-time forecasting method - Google Patents

Data-driven marine ship motion attitude real-time forecasting method Download PDF

Info

Publication number
CN113156815B
CN113156815B CN202110261215.9A CN202110261215A CN113156815B CN 113156815 B CN113156815 B CN 113156815B CN 202110261215 A CN202110261215 A CN 202110261215A CN 113156815 B CN113156815 B CN 113156815B
Authority
CN
China
Prior art keywords
ship
motion
data
yaw
information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110261215.9A
Other languages
Chinese (zh)
Other versions
CN113156815A (en
Inventor
尹建川
王宁
贾宝柱
潘新祥
徐进
廖志强
孔德峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Ocean University
Original Assignee
Guangdong Ocean University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong Ocean University filed Critical Guangdong Ocean University
Priority to CN202110261215.9A priority Critical patent/CN113156815B/en
Publication of CN113156815A publication Critical patent/CN113156815A/en
Application granted granted Critical
Publication of CN113156815B publication Critical patent/CN113156815B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a data-driven marine ship motion attitude real-time forecasting method, which is used for carrying out information screening, information decomposition, information processing and information fusion on ship six-degree-of-freedom motion information, and establishing a ship motion forecasting model based on multi-dimensional data to forecast the future ship motion attitudeThe six-degree-of-freedom motion is forecasted, so that the comprehensive ship motion attitude forecast is carried out. The invention is about the roll angle
Figure DDA0002970072530000011
Directly forecasting a longitudinal rocking angle theta and a yawing angle psi; for the longitudinal position X, the transverse position Y and the yaw angle psi, the surge velocity u, the surge velocity v and the yaw angular velocity r are firstly forecasted, and then forecast values of the longitudinal position X, the transverse position Y and the yaw angle psi are obtained. The method fully utilizes the actually measured ship motion state time sequence information, and can obtain more accurate ship motion forecast at sea by information mining and time sequence forecast of the information and utilizing the nonlinear fitting capability of the radial basis function neural network.

Description

Data-driven marine ship motion attitude real-time forecasting method
Technical Field
The invention relates to a ship motion attitude forecasting technology, in particular to a data-driven marine ship motion attitude real-time forecasting method.
Background
The ship motion forecasting method based on the artificial neural network is characterized in that historical ship motion data in a past period of time are used as input data of the network, the threshold value of each neuron and the connection weight value among all layers of the network are optimized through repeated training of the network, and therefore knowledge stored in network learning is closer to the output (forecasting output) of an actual system when used for reasoning, namely an approximate forecasting value is obtained for new network input data. The method has the greatest advantage that the network can autonomously find out the mapping rule of the given sample through training, the processes of data analysis and modeling are omitted, and the problem processing is avoided. The neural network theory is applied and developed very rapidly, and has wider application prospect due to the nonlinear fitting capability and no need of making assumptions on a system model. However, the traditional neural network model adopts an off-line modeling mode, has no capability of updating data and models in real time, and cannot reflect the dynamic time-varying marine vessel motion characteristics.
The traditional ship motion forecast based on the ship motion mathematical model has the following problems:
(1) it is difficult to obtain an accurate mathematical model of the vessel's motion at sea. The ship movement is interfered by natural environments such as wind, wave and flow, the pushing and control forces of the propeller and the rudder, and the changes of ship movement characteristics caused by environmental load and the changes of the ship self load make it difficult to establish an accurate ship movement mathematical model for the ship movement prediction at sea.
(2) The model cannot reflect the influence of time-varying factors on the motion of the ship. Dynamic changes of time-varying factors such as hydrology (waves, ocean currents, tides, tidal currents, temperature, salinity and density) and weather (wind power, wind direction, air temperature, air pressure and precipitation) in the sea area where the ship is located influence the precision of the fixed structure forecasting model for forecasting the ship motion, and especially the forecasting error is large under the condition that the hydrological factors change violently.
(3) And a comprehensive offshore ship six-degree-of-freedom motion online forecasting model is lacked. The six-degree-of-freedom ship motion prediction model has practical requirements on safe operation and control of ships on the sea, and the prior research lacks a comprehensive and universal six-degree-of-freedom motion online prediction model.
In a word, the real-time prediction of the marine attitude of the ship plays an important role in ship steering in high storms, taking off and landing of an aircraft on a deck, launching and recovering of an underwater vehicle, maritime refuting, cable launching and recovering and other maritime construction operations. However, when a ship sails on the sea, the ship is influenced by environmental elements such as wind, waves and currents, the swaying motion of the ship has complex characteristics such as nonlinearity and dynamic time variation, and the forecasting difficulty is increased.
Disclosure of Invention
In order to solve the problems in the prior art, the invention aims to design a data-driven marine ship motion attitude real-time forecasting method of a six-degree-of-freedom motion online forecasting model, which can reflect the influence of time-varying factors on ship motion and has universality.
In order to achieve the above purpose, the basic idea of the invention is as follows: the information of the ship six-degree-of-freedom motion is subjected to information screening, information decomposition, information processing and information fusion, and a ship motion prediction model based on multi-dimensional data is established to predict the future six-degree-of-freedom motion of the ship, so that the comprehensive ship motion attitude prediction is performed. The invention is about the roll angle
Figure GDA0003578046400000021
Directly forecasting a longitudinal rocking angle theta and a yawing angle psi; for the longitudinal position X, the transverse position Y and the yaw angle psi, the surge velocity u, the surge velocity v and the yaw angular velocity r are firstly forecasted, and then forecast values of the longitudinal position X, the transverse position Y and the yaw angle psi are obtained.
The technical scheme of the invention is as follows: a data-driven marine vessel motion attitude real-time forecasting method comprises the following steps:
A. collecting data
Acquiring six-degree-of-freedom motion element data of a ship as standby information, wherein the six-degree-of-freedom motion element data of the ship comprise six-degree-of-freedom attitude information of the ship and ship speed information, and the six-degree-of-freedom attitude information of the ship comprises a longitudinal position X, a transverse position Y, a vertical position Z and a roll angle of the ship under a fixed coordinate system
Figure GDA0003578046400000022
A pitch angle θ and a yaw angle ψ; the ship speed information comprises a surging speed mu, a surging speed nu and a yawing angular speed r of the ship under a motion coordinate system.
B. Decomposing and information screening multi-scale wavelet
The roll angle in the six-freedom-degree motion element data of the ship
Figure GDA0003578046400000031
And respectively carrying out wavelet decomposition on six time sequences of the pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v and the yaw angular velocity r to respectively obtain respective approximate components and detail components. I.e. by wavelet decomposition, saidEach time sequence of the six time sequences is decomposed into a plurality of subsequences as S (t) respectively, and comprises approximate components
Figure GDA0003578046400000032
Subsequences and detail components
Figure GDA0003578046400000033
A subsequence.
Using roll angle in ship six-freedom motion element data
Figure GDA0003578046400000034
Using one variable of the pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v or the yaw velocity r as a forecast object, and analyzing the yaw angle by a correlation analysis method
Figure GDA0003578046400000035
Carrying out relevance sorting and information screening on approximate components and detail components of the pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v and the yaw angular velocity r, selecting the approximate components and the detail components with strong relevance with a forecast object as the input of a ship motion forecast model, and respectively establishing a ship motion forecast model by using the pitch angle phi and the detail components
Figure GDA0003578046400000036
The pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v and the yaw angular velocity r are used as ship motion prediction models of prediction objects, and the total number is six ship motion prediction models.
C. Ship motion prediction model based on radial basis function neural network
C1, establishing a sliding data window
The motion of the ship on the sea has the characteristic of dynamic time variation, in order to reflect the latest ship motion state, a sliding data window is established to observe the state of the ship motion, and a ship motion prediction model based on a radial basis function neural network is dynamically adjusted by utilizing input and output data updated in real time;
the sliding window is a fixed width FIFOAnd a data sample sequence, wherein after a new group of input-output data is received, the new data group is added into the sliding window, and the earliest group of data is moved out of the sliding window. Sliding window W at time tSDExpressed as:
WSD＝[(pt-L+1,qt-L+1),…,(pt,qt)],
wherein L is the width of the sliding window; the real-time dynamics of the mapping relationship is represented by the input-output data sets within the sliding window, i.e. by the input matrix P and the corresponding output vector Q, respectively:
Figure GDA0003578046400000041
Q＝[qt-L+1,…,qt]
in the formula, npIs the dimension of the input matrix.
And respectively using the input matrix P and the corresponding output vector Q as the input and the output of the radial basis function neural network to train and dynamically adjust the radial basis function neural network.
C2, after receiving new data samples in each step, updating the sliding data window, adding the newest samples to the window and deleting the oldest samples from the window. And directly adding a new data sample into the hidden layer to be used as a new hidden node.
C3, calculating a response matrix phi of the hidden layer, wherein:
Figure GDA0003578046400000042
wherein, cjIs the center of the jth hidden node, piFor the ith sample, | | · | | represents the euclidean distance, and σ is the width of the basis function; m is the number of hidden nodes. Carrying out orthogonal decomposition on the vector in the response matrix phi into phi WA by using Gram-Schmidt law to obtain W-W1,…,wM]And A ═ a1,…,aM]。
The error reduction rate is calculated as follows:
Figure GDA0003578046400000043
error reduction rate is normalized as follows;
Figure GDA0003578046400000051
c4, selecting those hidden nodes whose sum of contributions to the output is less than the set value. Until the sum of the normalized error reduction rates of the selected hidden nodes satisfies the following formula:
Figure GDA0003578046400000052
where ρ is the precision threshold. Selecting 1, 2.. g hidden nodes to form a set of preparatory deletion hidden nodes as follows:
Sk＝{c1,…,cg}
wherein, c1,…,cgRespectively, among which the 1 st to g-th preliminary deletion hidden nodes.
Taking successive M in the pastSStep the intersection I of the selected hidden node set and delete the hidden nodes in I as follows:
Figure GDA0003578046400000053
wherein S isk、Sk-1、
Figure GDA0003578046400000054
Are respectively the k, k-1 and k-MS+1 set of selected hidden nodes.
And C5, after each hidden node is determined, updating the connection weight from the hidden layer to the output layer.
The obtained connection weight from the response matrix phi of the radial basis function neural network to the output vector Q is obtained by a partial least square method.
Performing partial least squares regression operation between the response matrix phi and the output vector Q; after the principal component matrix T is extracted, the response matrix phi and the output vector Q are respectively projected on the principal component matrix T, and the radial basis function neural network based on partial least squares regression is obtained as follows:
Q＝TR+F＝ΦWR+F
wherein T is a principal component matrix of phi; w is a transformation matrix of phi; r is a regression coefficient matrix; f is the residual matrix.
D. Forecasting of parameters of ship motion
For the motion with shorter three periods of rolling, pitching and heaving motion, the roll angle
Figure GDA0003578046400000061
Directly forecasting a pitch angle theta and a yaw angle psi; and the longitudinal position, the transverse position and the yawing angle change relatively slowly, the surging speed u, the swaying speed v and the yawing angular speed r are firstly forecasted, and then the forecast values of the longitudinal position, the transverse position and the yawing angle are obtained on the basis of the surging speed u, the swaying speed v and the yawing angular speed r, and the specific formula is as follows:
Figure GDA0003578046400000062
further, the correlation analysis method comprises a partial correlation coefficient calculation method, a correlation coefficient calculation method and a regression analysis method.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention fully utilizes the actually measured time sequence information of the ship motion state, and the information reflects the influence of the hydrological meteorological factors of the water area where the ship is positioned on the ship motion and reflects the time-varying characteristic of the ship motion dynamics. By information mining and time sequence forecasting of the information and utilizing the nonlinear fitting capacity of the radial basis function neural network, more accurate marine ship motion forecasting can be obtained. In the traditional forecasting method based on the mechanism model, the established mechanism model is difficult to reflect the complex influence of time-varying aperiodic hydrological weather and other factors, and cannot reflect the time-varying characteristic of ship motion dynamics, so that the forecasting precision is not high, and a larger forecasting error is easy to occur under the condition of large influence of external environmental factors.
2. The invention does not need to measure environmental factors such as wind, wave and the like, only needs the measurement data of the self motion state of the ship, and is simple and easy to implement.
3. The invention belongs to data-driven ship marine motion forecast, does not depend on a ship motion mathematical model, and has strong universality.
4. The method is a six-degree-of-freedom marine ship motion forecast, can comprehensively evaluate the motion state of a ship through the six-degree-of-freedom forecast, and can determine the time of marine operation and plan future operation comprehensively according to the forecast, such as take-off and landing of an aircraft on a deck, retraction and release of an underwater vehicle, overconnection on the sea, cable retraction and release, ship steering in heavy storms and the like.
Drawings
Fig. 1 is a six-degree-of-freedom motion diagram of a ship according to the invention.
FIG. 2 is a flow chart of neural network model training of the present invention.
Fig. 3 is a flow chart of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The flow of a data-driven marine vessel motion attitude real-time forecasting method is shown in fig. 3, a fixed coordinate system and a hull coordinate system in step a are shown in fig. 1, and a specific example of decomposing and information screening multi-scale wavelets in step B is shown in fig. 2. Fig. 2 illustrates a ship rolling motion prediction model. The alternative input being an approximate component of the roll angle
Figure GDA0003578046400000071
And K detail components
Figure GDA0003578046400000072
Approximate component of pitch angle
Figure GDA0003578046400000073
And K detail components
Figure GDA0003578046400000074
Approximate component of heave
Figure GDA0003578046400000075
And K detail components
Figure GDA0003578046400000076
Approximate component of surge velocity
Figure GDA0003578046400000077
And K detail components
Figure GDA0003578046400000078
Approximate component of the yaw rate
Figure GDA0003578046400000079
And K detail components
Figure GDA00035780464000000710
Approximate component of yaw rate
Figure GDA00035780464000000711
And K detail components
Figure GDA00035780464000000712
And selecting component information with strong output correlation from the alternative input as the input of the model.
Taking d-step motion forecast model modeling of ship roll angle as an example, selecting the model according to the magnitude of partial correlation coefficient in the input of alternative information:
Figure GDA00035780464000000713
as an input to the d-step motion prediction model,
Figure GDA00035780464000000714
and training the radial basis function neural network by taking the output of the d-step motion prediction model as a training sample, and establishing the d-step motion prediction model of ship rolling.
After a ship motion forecast model is built by utilizing the radial basis function neural network in the step C, the ship motion forecast model is used for
Figure GDA0003578046400000081
In order to be an input, the user can select,
Figure GDA0003578046400000082
and d-step advanced prediction is carried out for the output of the ship motion prediction model.
The present invention is not limited to the embodiment, and any equivalent idea or change within the technical scope of the present invention is to be regarded as the protection scope of the present invention.

Claims (2)

1. A data-driven marine vessel motion attitude real-time forecasting method is characterized by comprising the following steps: the method comprises the following steps:
A. collecting data
Acquiring six-degree-of-freedom motion element data of a ship as standby information, wherein the six-degree-of-freedom motion element data of the ship comprise six-degree-of-freedom attitude information of the ship and ship speed information, and the six-degree-of-freedom attitude information of the ship comprises a longitudinal position X, a transverse position Y, a vertical position Z and a roll angle of the ship under a fixed coordinate system
Figure FDA0003578046390000017
A pitch angle θ and a yaw angle ψ; the ship speed information comprises a surging speed mu, a surging speed nu and a yawing angular speed r of the ship under a motion coordinate system;
B. decomposing and information screening multi-scale wavelet
The roll angle in the six-freedom-degree motion element data of the ship
Figure FDA0003578046390000011
Carrying out wavelet decomposition on six time sequences of a pitch angle theta, a yaw angle psi, a pitch velocity u, a yaw velocity v and a yaw angular velocity r respectively to obtain respective approximate components and detail components; namely, decomposing each time sequence of the six time sequences into a plurality of subsequences respectively as S (t) through wavelet decomposition, wherein each subsequence comprises approximate components
Figure FDA0003578046390000012
Subsequences and detail components
Figure FDA0003578046390000013
A subsequence;
roll angle in ship six-freedom-degree motion element data
Figure FDA0003578046390000014
Using one variable of the pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v or the yaw velocity r as a forecast object, and analyzing the yaw angle by a correlation analysis method
Figure FDA0003578046390000015
Carrying out relevance sorting and information screening on approximate components and detail components of the pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v and the yaw angular velocity r, selecting the approximate components and the detail components with strong relevance with a forecast object as the input of a ship motion forecast model, and respectively establishing a ship motion forecast model by using the pitch angle phi and the detail components
Figure FDA0003578046390000016
The pitch angle theta, the yaw angle psi, the pitch velocity u, the yaw velocity v and the yaw angular velocity r are used as ship motion prediction models of prediction objects, and the total number of the ship motion prediction models is six;
C. ship motion forecasting model based on radial basis function neural network
C1, establishing a sliding data window
The motion of the ship on the sea has the characteristic of dynamic time variation, in order to reflect the latest ship motion state, a sliding data window is established to observe the state of the ship motion, and a ship motion prediction model based on a radial basis function neural network is dynamically adjusted by utilizing input and output data updated in real time;
the sliding window is a first-in first-out data sample sequence with fixed width, when a new group of input-output data is received, the new data group is added into the sliding window, and the earliest group of data is shifted out of the sliding window; sliding window W at time tSDExpressed as:
WSD＝[(pt-L+1,qt-L+1),…,(pt,qt)],
wherein L is the width of the sliding window; the real-time dynamics of the mapping relationship is represented by the input-output data sets within the sliding window, i.e. by the input matrix P and the corresponding output vector Q, respectively:
Figure FDA0003578046390000021
Q＝[qt-L+1,…,qt]
in the formula, npIs the dimension of the input matrix;
respectively using the input matrix P and the corresponding output vector Q as the input and the output of the radial basis function neural network, and training and dynamically adjusting the radial basis function neural network;
c2, after receiving new data sample in each step, updating the sliding data window, adding the newest sample into the window, and deleting the oldest sample from the window; directly adding a new data sample into the hidden layer to be used as a new hidden node;
c3, calculating a response matrix phi of the hidden layer, wherein:
Figure FDA0003578046390000031
wherein, cjIs the center of the jth hidden node, piFor the ith sample, | | · | | represents the euclidean distance, and σ is the width of the basis function; m is the number of hidden nodes; carrying out orthogonal decomposition on the vector in the response matrix phi into phi WA by using Gram-Schmidt law to obtain W-W1,…,wM]And A ═ a1,…,aM]；
The error reduction rate is calculated as follows:
Figure FDA0003578046390000032
error reduction rate is normalized as follows;
Figure FDA0003578046390000033
c4, selecting those hidden nodes whose sum of contributions to the output is less than the set value; until the sum of the normalized error reduction rates of the selected hidden nodes satisfies the following formula:
Figure FDA0003578046390000034
wherein rho is a precision threshold; selecting 1, 2.. g hidden nodes to form a set of preparatory deletion hidden nodes as follows:
Sk＝{c1,…,cg}
wherein, c1,…,cgRespectively, the hidden nodes of the 1 st to the g th preparatory deletion;
taking successive M in the pastSStep the intersection I of the selected hidden node set and delete the hidden nodes in I as follows:
Figure FDA0003578046390000035
wherein S isk、Sk-1、
Figure FDA0003578046390000036
Are respectively the k, k-1 and k-MS+1, a set of selected hidden nodes;
c5, after each hidden node is determined, updating the connection weight from the hidden layer to the output layer;
the obtained connection weight from the response matrix phi of the radial basis function neural network to the output vector Q is obtained by a partial least square method;
performing partial least squares regression operation between the response matrix phi and the output vector Q; after the principal component matrix T is extracted, the response matrix phi and the output vector Q are respectively projected on the principal component matrix T, and the radial basis function neural network based on partial least squares regression is obtained as follows:
Q＝TR+F＝ΦWR+F
wherein T is a principal component matrix of phi; w is a transformation matrix of phi; r is a regression coefficient matrix; f is a residual error matrix;
D. forecasting of parameters of ship motion
For the motion with shorter three periods of rolling, pitching and heaving motion, the roll angle
Figure FDA0003578046390000042
Directly forecasting a pitch angle theta and a yaw angle psi; and the longitudinal position, the transverse position and the yawing angle change relatively slowly, the surging speed u, the swaying speed v and the yawing angular speed r are firstly forecasted, and then the forecast values of the longitudinal position, the transverse position and the yawing angle are obtained on the basis of the surging speed u, the swaying speed v and the yawing angular speed r, and the specific formula is as follows:
Figure FDA0003578046390000041
2. the data-driven marine vessel motion attitude real-time forecasting method as claimed in claim 1, characterized in that: the correlation analysis method comprises a partial correlation coefficient calculation method, a correlation coefficient calculation method and a regression analysis method.
CN202110261215.9A 2021-03-10 2021-03-10 Data-driven marine ship motion attitude real-time forecasting method Active CN113156815B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110261215.9A CN113156815B (en) 2021-03-10 2021-03-10 Data-driven marine ship motion attitude real-time forecasting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110261215.9A CN113156815B (en) 2021-03-10 2021-03-10 Data-driven marine ship motion attitude real-time forecasting method

Publications (2)

Publication Number Publication Date
CN113156815A CN113156815A (en) 2021-07-23
CN113156815B true CN113156815B (en) 2022-05-20

Family

ID=76886607

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110261215.9A Active CN113156815B (en) 2021-03-10 2021-03-10 Data-driven marine ship motion attitude real-time forecasting method

Country Status (1)

Country Link
CN (1) CN113156815B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113837454B (en) * 2021-09-09 2024-04-23 武汉大学 Ship three-degree-of-freedom hybrid neural network model prediction method and system
CN116702095B (en) * 2023-06-01 2024-04-05 大连海事大学 Modularized marine ship motion attitude real-time forecasting method
CN116819950B (en) * 2023-08-25 2023-11-07 中国海洋大学 Ship and floating ocean platform dynamic positioning control method and system
CN117518298A (en) * 2023-12-29 2024-02-06 大连海事大学 Tidal combination forecasting method with self-adaptive model structure
CN117494057A (en) * 2023-12-29 2024-02-02 大连海事大学 Ship motion prediction method based on hierarchical multi-scale decomposition

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7756689B1 (en) * 2006-02-28 2010-07-13 The United States Of America As Represented By The Secretary Of The Navy Numerical modeling of six-degree-freedom ship motion
CN108846207B (en) * 2018-06-15 2023-08-01 哈尔滨工程大学 Ship motion forecasting device and forecasting method based on nonlinear model identification
CN110032068A (en) * 2019-04-02 2019-07-19 杭州国辰正域科技有限公司 A kind of online Ship Movement Prediction and Anti-roll method and system

Also Published As

Publication number Publication date
CN113156815A (en) 2021-07-23

Similar Documents

Publication Publication Date Title
CN113156815B (en) Data-driven marine ship motion attitude real-time forecasting method
CN112684701A (en) Ship motion forecasting method based on long-time and short-time memory network and Gaussian process regression
CN112116162B (en) Power transmission line icing thickness prediction method based on CEEMDAN-QFAOA-LSTM
Kawai et al. Sea state estimation using monitoring data by convolutional neural network (CNN)
CN114021441A (en) Ship motion attitude prediction method based on CNN-BilSTM
CN116702095B (en) Modularized marine ship motion attitude real-time forecasting method
CN105836066A (en) Method and system for predicting posture of ship moored at open type wharf
Mittendorf et al. Sea state identification using machine learning—A comparative study based on in-service data from a container vessel
CN110737267A (en) Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships
Zhang et al. Trajectory prediction based on AIS and BP neural network
Wu et al. A modular tide level prediction method based on a NARX neural network
CN113837454B (en) Ship three-degree-of-freedom hybrid neural network model prediction method and system
CN107122897B (en) Unmanned ship airworthiness evaluation method based on deep learning algorithm
Zhang et al. Model identification of ship turning maneuver and extreme short-term trajectory prediction under the influence of sea currents
Marie et al. Sail-assisted motor vessels weather routing using a fuzzy logic model
Gupta et al. Big data analytics as a tool to monitor hydrodynamic performance of a ship
CN115410419B (en) Ship mooring early warning method, system, electronic equipment and storage medium
CN116127846A (en) Intelligent evaluation method and system for anti-rolling technology
Mei et al. Ship maneuvering prediction using grey box framework via adaptive RM-SVM with minor rudder
Huang et al. Short-term prediction of ship pitching motion based on artificial neural networks
Domala et al. A Univariate and multivariate machine learning approach for prediction of significant wave height
Liu et al. Improving deterministic pitch motions estimation using bivariate sequential wave input
Zhou et al. Nonparametric modeling of ship maneuvering motions in calm water and regular waves based on R-LSTM hybrid method
CN117494057A (en) Ship motion prediction method based on hierarchical multi-scale decomposition
Yao et al. Extreme motion prediction and early-warning assessment of semisubmersible platform based on deep learning method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant