Research On Wave Spectrum Estimation And Wind Surge Separation Technology On Wave Buoy

With the continuous development and expansion of China's ocean power,ocean observation has played a more and more important role in the construction of China's Marine power.Wave spectrum is an important statistical feature of random waves,which is of great significance for ocean observation.The waves on the ocean surface are composed of wind waves and surge waves in most of the time and region.The internal properties and external appearance of wind waves and surge waves are very different,and they also have completely different influences on the scientific research and ship navigation at sea.Therefore,the research on wave spectrum estimation and wind surge separation is of great significance.Therefore,based on the measured wave data of SZF wave buoys obtained by national defense scientific research project,the wave spectrum estimation and wind surge separation technology of wave buoys are studied in this paper.In this paper,after consulting a large number of domestic and foreign papers,it is found that most of the researches on wave spectrum estimation are carried out by using the classical spectral estimation method,and almost few modern wave spectrum estimation methods are used.Therefore,in this paper,the wave data collected by wave buoys are used to study the wave power spectrum estimation and direction spectrum estimation using the AR model method and the extended eigenvector method.In order to realize the separation of wind and surge and select a more suitable for engineering application of wind surge separation method.In this paper,four wind surge separation methods with surface characteristics are used to conduct experiments by using the estimated wave spectrum,and a more suitable wind surge separation method is obtained by comparing and analyzing the separation results.Firstly,this paper uses the 3? outlier detection principle and the weighted outlier substitution method to preprocess the wave data collected by the buoy.The AR model method is proposed to estimate the wave power spectrum from the processed data and get the wave power spectrum with good performance.The extended eigenvector method is used to invert the wave direction spectrum,and the wave direction spectrum with good performance is obtained.The estimated wave spectrum is used to calculate the wave elements.Secondly,the principle and separation steps of several typical wind surge separation methods,such as spectral integration method,fixed frequency method,OP method and PM method,are introduced in detail.Under the mixed sea conditions of pure wind wave and surge wave,the estimated wave power spectrum is processed by using the above four methods,and the separation performance of the four methods is tested by using the trend diagram of wave direction and frequency as the reference.The results show that the frequency spectrum integral method under the buoy platform can achieve good results in wind surge separation,and is more suitable for engineering applications.Finally,the PM spectrum model is used as the real target spectrum to measure the spectrum estimation quality of the estimated wave power spectrum.The results show that the AR model method can be used to estimate the wave power spectrum with good performance.The wave elements obtained by the two dimensional spectral wind surge separation experiment are compared with the spectrum integral method,and the results show that their correlation is good.A large number of experimental comparisons show that satisfactory separation results can be obtained by using the one-dimensional spectrum integral method.