CN-122020029-A - Data processing method and device for ship attitude prediction

CN122020029ACN 122020029 ACN122020029 ACN 122020029ACN-122020029-A

Abstract

The application discloses a data processing method and device for ship attitude prediction. The method comprises the steps of obtaining ship motion data to be processed, wherein the ship motion data to be processed are data used for representing ship motion, carrying out feature extraction processing based on multidimensional fusion on the ship motion data to be processed to obtain multidimensional fusion motion feature data, wherein the multidimensional fusion motion feature data are feature data used for representing ship motion multidimensional feature fusion, and carrying out gesture prediction processing based on a ship gesture prediction model on the multidimensional fusion motion feature data to obtain predicted ship gesture data. By carrying out ship attitude prediction based on multidimensional fusion on ship motion data, the problem that in the prior art, the accuracy rate of ship attitude prediction is low is solved, and the technical effect of improving the accuracy rate of ship attitude prediction is realized.

Inventors

DONG WENTAO
CHEN TONG
Luan Xinrui
LIU YING
YIN FEI
ZHANG YIMENG
DONG YUCAI
ZHANG XIAOWEI
Kong Zining
XIAO LONGBIN
WANG SHENGXU
WANG QIANG
CUI WEI
LIN YUANYUAN
ZHANG SHITAI

Assignees

中国电子科技集团公司第十五研究所

Dates

Publication Date: 20260512
Application Date: 20260114

Claims (10)

1. A data processing method for hull attitude prediction, comprising: Acquiring ship motion data to be processed, wherein the ship motion data to be processed is data for representing ship shaking motion; Performing feature extraction processing based on multidimensional fusion on the ship motion data to be processed to obtain multidimensional fusion motion feature data, wherein the multidimensional fusion motion feature data is feature data for representing multidimensional feature fusion of ship motion; And carrying out attitude prediction processing based on a ship attitude prediction model on the multidimensional fusion motion characteristic data to obtain predicted ship attitude data.
2. The data processing method according to claim 1, wherein performing feature extraction processing based on multidimensional fusion on the hull motion data to be processed, obtaining multidimensional fusion motion feature data includes: Performing multidimensional feature extraction processing on the ship motion data to be processed to obtain multidimensional motion feature data, wherein the multidimensional motion feature data is data for representing multidimensional features of ship motion; Performing multidimensional coupling feature extraction on the ship movement data to be processed to obtain coupling feature data, wherein the coupling feature data are feature data used for representing coupling among ship movement multidimensional features; and carrying out feature fusion processing on the multidimensional movement feature data based on the coupling feature data to obtain the multidimensional fusion movement feature data.
3. The data processing method according to claim 2, wherein performing multidimensional feature extraction processing on the hull motion data to be processed, obtaining multidimensional motion feature data includes; extracting the ship motion data to be processed based on a first frequency characteristic to obtain first motion characteristic data, wherein the first motion characteristic data is characteristic data used for representing low frequency of a ship motion period; Extracting the ship motion data to be processed based on a second frequency characteristic to obtain second motion characteristic data, wherein the second motion characteristic data is characteristic data for representing high-frequency shaking of ship motion; Performing feature extraction processing based on local response on the ship motion data to be processed to obtain third motion feature data, wherein the third motion feature data is feature data for representing sudden deflection of ship motion; And obtaining the multidimensional motion characteristic data according to the first motion characteristic data, the second motion characteristic data and the third motion characteristic data.
4. The data processing method according to claim 2, wherein the multidimensional coupling feature extraction is performed on the ship movement data to be processed, and obtaining coupling feature data includes: extracting the ship rolling motion data to be processed based on rolling characteristics to obtain rolling characteristic data, wherein the rolling characteristic data are characteristic data used for representing ship rolling motion; extracting the ship pitching motion data to be processed based on pitching characteristics to obtain pitching characteristic data, wherein the pitching characteristic data is characteristic data for representing pitching motion of a ship body; extracting the ship swaying motion data to be processed based on heave characteristics to obtain heave characteristic data, wherein the heave characteristic data is characteristic data for representing ship swaying motion; and performing characteristic-based coupling processing on the rolling characteristic data, the pitching characteristic data and the heave characteristic data to obtain the coupling characteristic data.
5. The data processing method according to claim 4, wherein performing feature-based coupling processing on the roll feature data, the pitch feature data, and the heave feature data to obtain the coupling feature data includes: Performing feature coupling processing on the rolling feature data and the pitching feature data based on element-by-element products to obtain associated feature data, wherein the associated feature data is feature data used for representing interaction between the rolling feature and the pitching feature; performing characteristic coupling processing based on dynamic weighting on the rolling characteristic data, the pitching characteristic data and the heave characteristic data to obtain linear superposition characteristic data, wherein the linear superposition characteristic data is characteristic data used for representing linear superposition among the rolling characteristic, the pitching characteristic and the heave characteristic; And carrying out feature fusion processing on the associated feature data and the linear superposition feature data to obtain the coupling feature data.
6. The data processing method according to claim 1, wherein performing attitude prediction processing based on a hull attitude prediction model on the multidimensional fusion motion feature data to obtain predicted hull attitude data includes: performing ship-shake state prediction based on time-step mapping on the multidimensional fusion motion characteristic data to obtain ship-shake prediction data; And carrying out gesture prediction processing based on preset equipment on the ship shaking prediction data to obtain predicted ship body gesture data.
7. A data processing method according to claim 1, wherein prior to acquiring hull motion data to be processed, the method further comprises: acquiring sample hull motion data, wherein the sample hull motion data is sample data for representing ship movement; Performing feature extraction processing based on multidimensional fusion on the sample ship body motion data to obtain sample multidimensional fusion motion feature data; And carrying out prediction model construction processing on the sample multidimensional fusion motion characteristic data to obtain the ship attitude prediction model.
8. A data processing apparatus for predicting the attitude of a ship hull, comprising: The data acquisition module is used for acquiring ship movement data to be processed, wherein the ship movement data to be processed is data for representing ship rocking movement; The multidimensional feature module is used for carrying out feature extraction processing based on multidimensional fusion on the ship motion data to be processed to obtain multidimensional fusion motion feature data, wherein the multidimensional fusion motion feature data is feature data used for representing multidimensional feature fusion of ship motion; And the prediction module is used for carrying out attitude prediction processing based on a ship attitude prediction model on the multidimensional fusion motion characteristic data to obtain predicted ship attitude data.
9. A computer-readable storage medium storing computer instructions for causing the computer to execute the data processing method for hull attitude prediction according to any one of claims 1 to 7.
10. An electronic device comprising at least one processor and a memory communicatively coupled to the at least one processor, wherein the memory stores a computer program executable by the at least one processor to cause the at least one processor to perform the data processing method for hull attitude prediction of any of claims 1-7.

Description

Data processing method and device for ship attitude prediction Technical Field The application relates to the field of computers, in particular to a data processing method and device for ship attitude prediction. Background The survey vessel assumes the task of tracking and surveying aircraft targets under sea-based conditions. Under sea-based conditions, the survey vessel is affected by wind, waves, gushes, etc., and inevitably shakes, so that the survey reference-deck plane of the survey vessel is inclined. If the real-time prediction of the ship swinging motion at the present time output by the inertial navigation device on the survey ship can be performed, the influence of the ship body posture of the survey ship can be isolated when the track data is used for digitally guiding the measurement and control device, so that the measurement and control device can be used for better tracking and capturing the flying target. Therefore, it is necessary to study the ocean going survey vessel hull attitude data prediction model. The traditional time sequence modeling method, such as autoregressive moving average model (ARMA) and Legend polynomial fitting method, has blindness and randomness in the aspects of difficulty in constructing a ship motion state equation and related parameter setting according to the task characteristics of a measuring ship, so that the ship attitude prediction accuracy is low. Therefore, the application is provided for solving the problem of lower accuracy rate of ship attitude prediction in the prior art. Disclosure of Invention The application mainly aims to provide a data processing method and device for predicting the ship body posture, which are used for solving the problem of lower accuracy rate of the ship body posture prediction in the prior art and realizing the technical effect of improving the accuracy rate of the ship body posture prediction. To achieve the above object, a first aspect of the present application proposes a data processing method for hull attitude prediction, including: Acquiring ship motion data to be processed, wherein the ship motion data to be processed is data for representing ship shaking motion; Performing feature extraction processing based on multidimensional fusion on the ship motion data to be processed to obtain multidimensional fusion motion feature data, wherein the multidimensional fusion motion feature data is feature data for representing multidimensional feature fusion of ship motion; And carrying out attitude prediction processing based on a ship attitude prediction model on the multidimensional fusion motion characteristic data to obtain predicted ship attitude data. Further, performing feature extraction processing based on multidimensional fusion on the ship motion data to be processed, and obtaining multidimensional fusion motion feature data includes: Performing multidimensional feature extraction processing on the ship motion data to be processed to obtain multidimensional motion feature data, wherein the multidimensional motion feature data is data for representing multidimensional features of ship motion; Performing multidimensional coupling feature extraction on the ship movement data to be processed to obtain coupling feature data, wherein the coupling feature data are feature data used for representing coupling among ship movement multidimensional features; and carrying out feature fusion processing on the multidimensional movement feature data based on the coupling feature data to obtain the multidimensional fusion movement feature data. Further, multidimensional feature extraction processing is carried out on the ship motion data to be processed, and multidimensional motion feature data are obtained and comprise; extracting the ship motion data to be processed based on a first frequency characteristic to obtain first motion characteristic data, wherein the first motion characteristic data is characteristic data used for representing low frequency of a ship motion period; Extracting the ship motion data to be processed based on a second frequency characteristic to obtain second motion characteristic data, wherein the second motion characteristic data is characteristic data for representing high-frequency shaking of ship motion; Performing feature extraction processing based on local response on the ship motion data to be processed to obtain third motion feature data, wherein the third motion feature data is feature data for representing sudden deflection of ship motion; And obtaining the multidimensional motion characteristic data according to the first motion characteristic data, the second motion characteristic data and the third motion characteristic data. Further, performing multidimensional coupling feature extraction on the ship movement data to be processed, and obtaining coupling feature data includes: extracting the ship rolling motion data to be processed based on rolling characteristics to obtain rolling characterist