Research On Prediction Of The Significant Wave Height Based On Ship Attitude

The investigation of sea wave is one of the important issues in Naval Architecture and Ocean Engineering. As is known to all, wave is a complex fluctuate phenomenon resulting from many factors, such as wind force, atmosphere and ocean current, etc. It is also veiwed as a small-scale surface gravity wave because of wind. So far there is no perfect and rigorous theory to predict various aspects of wave because of the complexity. However, the prediction of sea waves has great practical value for application. It is very important for the safety of ships in the ocean, provided the waves are predicted accurately, so the investigation of sea wave is valued by the wave scientists.To predict the waves, we must get the accurate information on the waves. For the moment, a lot of method to obtain the wave information has been found, including: telemetry waves, wave imaging measurement system, onboard measurements and fixed-point measurement and etc. However, there are many limitations on the use of these methods, and make them can not be universally applied so the main subject is find more convenient and widely used method of wave forecast in the current wave forecast.This thesis is concerned with the prediction of significant wave height. In order to get rid of the limitations of using the measurement of wave information instrument, this thesis is intended to sail the ship as a wave instrument to measure the waves of information. The main content of this thesis is as follows:Firstly, this thesis summarizes the researches about sea wave height prediction at home and abroad and introduces some basic techniques and theory on the prediction of sea wave,then sum up these methods briefly. A kinematics and dynamics model about the ship is established, which is useful for further researches.Secondly, the hydrodynamic parameters and the states of ship are obtained by UKF, then the waves disturbance is deducd by equations of motion of the ship. As there is a certain relationship between the wave disturbance force and the sea wave height, here neural network is used to find the relationship. Specifically, wave force is the input, and wave height is the output. At the same time, comparing the predicted value with the true value to verify the proposed method.Thirdly, an algorithm called generalized dynamic fuzzy neural network is researched,which is based on extended radial basis neural networks. It uses a linear transformation to approximate any nonlinear system, so this algorithm is more general and has a high degree of accuracy than other neural network algorithm. The thesis apply this algorithm to forecast the significant wave height directly and can improve the forecasting accuracy and speed effectively.Finally, a comparison between effective wave height obtained from GD-FNN and wave height height instrument is made. It is verified about the application and reliabity by predicting wave height using ship parameters.