Identification-based simplified model of large container ships using support vector machines and artificial bee colony algorithm

The 6 degrees of freedom (DOF) model with a high degree of complexity for capturing ship dynamics is generally able to track the nonlinear and coupling dynamics of ships. However, the 6 DOF model makes challenges in estimating model coefficients and designing the model-based control. Therefore, simplified ship dynamic models within allowed accuracy are essential. This paper simplified the 6 DOF nonlinear dynamic model of ships into two decoupled models including the speed model and the steering model through reasonable assumptions. Those models were tested through maneuvering simulations of a container ship with a 4 DOF dynamic model. Support vector machines (SVM) optimized by the artificial bee colony algorithm (ABC) was used to identify parameters of speed and steering models by analyzing the rudder angle, propeller shaft speed, surge and sway velocities, and yaw rate from simulated data extracted from a series of maneuvers made by the container ship. Comparisons with the first order linear and nonlinear Nomoto models show that the simplified nonlinear steering model can capture more complicated dynamics and performs better. Additionally, comparisons among three different parameter identification methods demonstrate similar identification results but the different performance involving the applicability and effectiveness. SVM optimized by ABC is relatively convenient and effective for parameter identification of ship simplified dynamic models.     

支持向量回归的机器学习方法在海浪预测中的应用

基于支持向量回归(SVR)方法,建立了渤海海域近岸海浪有效波高短期预测模型,并设计了多组风浪信息组合输入方案,开展了有效波高预测敏感性试验。研究发现:综合考虑当前风浪信息作为模型的输入,对3 h和6 h有效波高预测具有较高的预报技巧,但随着预测时效的延长其预测准确性迅速降低;若此时引入未来预测风速信息作为模型输入,则可极大提高对12 h和24 h有效波高的预测能力;此外,若输入信息与预测对象之间不存在显著相关,多个信息的输入对有效波高预测效果提高无显著作用。建立的机器学习模型对小样本数据集具有良好的适应能力,能够有效解决海浪预报中的非线性问题,可为近岸海浪有效波高短期预测提供合理的技术参考。     

Study of tide prediction method influenced by nonperiodic factors based on support vector machines

Harmonic analysis, the traditional tidal forecasting method, cannot take into account the impact of noncyclical factors, and is also based on the BP neural network tidal prediction model which is easily limited by the amount of data. According to the movement of celestial bodies, and considering the insufficient tidal characteristics of historical data which are impacted by the nonperiodic weather, a tidal prediction method is designed based on support vector machine (SVM) to carry out the simulation experiment by using tidal data from Xiamen Tide Gauge, Luchaogang Tide Gauge and Weifang Tide Gauge individually. And the results show that the model satisfactorily carries out the tide prediction which is influenced by noncyclical factors. At the same time, it also proves that the proposed prediction method, which when compared with harmonic analysis method and the BP neural network method, has faster modeling speed, higher prediction precision and stronger generalization ability.     

Artificial neural network to translate offshore satellite wave data to coastal locations

Owing to the spatial averaging involved in satellite sensing, use of observations so collected is often restricted to offshore regions. This paper discusses a technique to obtain significant wave heights at a specified coastal site from their values gathered by a satellite at deeper offshore locations. The technique is based on the approach of Artificial Neural Network (ANN) of Radial Basis Function (RBF) and Feed-forward Back-propagation (FFBP) type. The satellite-sensed data of significant wave height; average wave period and the wind speed were given as input to the network in order to obtain significant wave heights at a coastal site situated along the west coast of India. Qualitative as well as quantitative comparison of the network output with target observations showed usefulness of the selected networks in such an application vis-à-vis simpler techniques like statistical regression. The basic FFBP network predicted the higher waves more correctly although such a network was less attractive from the point of overall accuracy. Unlike satellite observations collection of buoy data is costly and hence, it is generally resorted to fewer locations and for a smaller period of time. As shown in this study the network can be trained with samples of buoy data and can be further used for routine wave forecasting at coastal locations based on more permanent flow of satellite observations.     

Wave height forecasting in Dayyer, the Persian Gulf

Forecasting of wave parameters is necessary for many marine and coastal operations. Different forecasting methodologies have been developed using the wind and wave characteristics. In this paper, artificial neural network (ANN) as a robust data learning method is used to forecast the wave height for the next 3, 6, 12 and 24 h in the Persian Gulf. To determine the effective parameters, different models with various combinations of input parameters were considered. Parameters such as wind speed, direction and wave height of the previous 3 h, were found to be the best inputs. Furthermore, using the difference between wave and wind directions showed better performance. The results also indicated that if only the wind parameters are used as model inputs the accuracy of the forecasting increases as the time horizon increases up to 6 h. This can be due to the lower influence of previous wave heights on larger lead time forecasting and the existing lag between the wind and wave growth. It was also found that in short lead times, the forecasted wave heights primarily depend on the previous wave heights, while in larger lead times there is a greater dependence on previous wind speeds.     

On the choice of data transformation for modelling time series of significant wave height

Autoregressive models have been shown to adequately model the time series of significant wave height. However, since this series exhibits a seasonal component and has a non-gaussian nature, it is necessary to transform the series before a model can be fit to the data. Two different transformations that have been used in earlier work are shown not to be appropriate for all types of applications. A third transformation is proposed here, which combines the better features of the two earlier ones and which is appropriate for simulation work. This is demonstrated with an example of a series from Figueira da Foz, a location of the Portuguese Coast.     

Prediction of maximum wave-induced liquefaction in porous seabed using multi-artificial neural network model

In the last few decades, considerable efforts have been devoted to the phenomenon of wave-induced liquefactions, because it is one of the most important factors for analysing the seabed and designing marine structures. Although numerous studies of wave-induced liquefaction have been carried out, comparatively little is known about the impact of liquefaction on marine structures. Furthermore, most previous researches have focused on complicated mathematical theories and some laboratory work. In the present study, a data dependent approach for the prediction of the wave-induced liquefaction depth in a porous seabed is proposed, based on a multi-artificial neural network (MANN) method. Numerical results indicate that the MANN model can provide an accurate prediction of the wave-induced maximum liquefaction depth with 10% of the original database. This study demonstrates the capacity of the proposed MANN model and provides coastal engineers with another effective tool to analyse the stability of the marine sediment.     

Wave energy assessment based on trivariate distribution of significant wave height,mean period and direction

Based on the Vine copula theory, a trivariate statistical model of significant wave height, characterized wave period and mean wave direction was constructed. To maintain the properties of the different types of variables, a special copula function was derived from the model developed by Johnson and Wehrly based on the maximum entropy principle. It was then combined with the Archimedean copulas to construct the proposed model. An effective algorithm for generating corresponding joint pseudo-random numbers was also developed. Statistical analysis of hindcast data for the significant wave height, mean wave period, and direction, which were collected from an observation point in the North Atlantic every three hours from 1997 to 2001, was performed. The marginal distributions of the significant wave height and mean wave period were fitted by a modified maximum entropy distribution, and the mean wave direction was fitted by a mixture of von Mises distributions. It was shown that the proposed model is a good fit for the data. The seasonal wave energy resources in the target area were assessed using the model estimates. Histograms of the directional wave energy, wave energy roses, and scatter and energy diagrams were presented.     

Measures to diminish the parameter drift in the modeling of ship manoeuvring using system identification

System identification provides an effective way to predict the ship manoeuvrability. In this paper several measures are proposed to diminish the parameter drift in the parametric identification of ship manoeuvring models. The drift of linear hydrodynamic coefficients can be accounted for from the point of view of dynamic cancellation, while the drift of nonlinear hydrodynamic coefficients is explained from the point of view of regression analysis. To diminish the parameter drift, reconstruction of the samples and modification of the mathematical model of ship manoeuvring motion are carried out. Difference method and the method of additional excitation are proposed to reconstruct the samples. Using correlation analysis, the structure of a manoeuvring model is simplified. Combined with the measures proposed, support vector machines based identification is employed to determine the hydrodynamic coefficients in a modified Abkowitz model. Experimental data from the free-running model tests of a KVLCC2 ship are analyzed and the hydrodynamic coefficients are identified. Based on the regressive model, simulation of manoeuvres is conducted. Comparison between the simulation results and the experimental results demonstrates the validity of the proposed measures.     

Bivariate autoregressive models for the time series of significant wave height and mean period

This paper generalises the application of univariate models of the long-term time series of significant wave height to the case of the bivariate series of significant wave height and mean period. A brief review of the basic features of multivariate autoregressive models is presented, and then applications are made to the wave time series of Figueira da Foz, in Portugal. It is demonstrated that the simulated series from these models exhibit the correlation between the two parameters a feature that univariate series cannot reproduce. An application to two series of significant wave height from two neighbouring stations shows the applicability of this type of models to other type of correlated data sets.     

Predicting significant wave height off the northeast coast of the United States

To develop a simple method to predict the significant wave height, we analyze 18 years of hourly observations from 12 different buoys that are off the northeast coast of the United States. Water depths ranged from 19 to 4427 m for these moored buoys. We find that, on average, all of these buoys exhibit a region of constant wave height for 10-m wind speeds between 0 and 4 m s⁻¹. That wave height does, however, depend on water depth. For wind speeds above 4 m s^–1, the wave height increases as the square of the wind speed; but the multiplicative factor is again a function of water depth. We synthesize these results in a prediction scheme that yields the significant wave height from simple functions of water depth and 10-m wind speed for wind speeds up to 25 m s^–1.     

The influence of seasonality on estimating return values of significant wave height

A time-dependent generalized extreme value (GEV) model for monthly significant wave heights maxima is developed. The model is applied to several 3-hour time series from the Spanish buoy network. Monthly maxima show a clear non-stationary behavior within a year, suggesting that the location, scale and shape parameters of the GEV distribution can be parameterized using harmonic functions. To avoid a possible over-parameterization, an automatic selection model, based on the Akaike Information Criterion, is carried out. Results show that the non-stationary behavior of monthly maxima significant wave height is adequately modeled, drastically increasing the significance of the parameters involved and reducing the uncertainty in the return level estimation. The model provides new information to analyze the seasonal behavior of wave height extremes affecting different natural coastal processes.     

Prediction of extreme wave heights using neuro wavelet technique

Precise prediction of extreme wave heights is still an evading problem whether it is done using physics based modeling or by extensively used data driven technique of Artificial Neural Network (ANN). In the present paper, Neuro Wavelet Technique (NWT) is used specifically to explore the possibility of prediction of extreme events for five major hurricanes Katrina 2005, Dean 2007, Gustav 2008, Ike 2008, Irene 2011 at four locations (NDBC wave buoys stations)¹ namely; 42040, 42039, 41004, 41041 in the Gulf of Mexico. Neuro Wavelet Technique is employed by combining Discrete Wavelet Transform and Artificial Neural Networks. Discrete wavelet transform analyzes frequency of signal with respect to time at different scales. It decomposes time series into low (approximate) and high (detail) frequency components. The decomposition of approximate components (extreme events in the ocean wave series) can be carried out up to the desired multiple levels in order to provide relatively smooth varying amplitude series. This feature of wavelet transforms make it plausible for predicting extreme events with a better accuracy. In the present study third, fifth and seventh level of decompositions are used which facilitates 3 to 7 times filtering of low frequency events and seems to pay the dividend in the form of better prediction accuracy at extreme events. To develop these Neuro wavelet models to forecast the waves with lead times of 12 hr to 36 hr in advance, previously measured significant wave heights at same locations were used. The results were judged by wave plots, scatter plots and other error measures. From the results it can be concluded that the Neuro Wavelet Technique can be employed to solve the ever eluding problem of accurate forecasting of the extreme events.     

Seasonality and duration in extreme value distributions of significant wave height

This paper presents a statistical model to characterize the long-term extreme value distribution of significant wave height, conditioning to the duration of the storm and accounting for seasonality. A time-dependent version of the peak over threshold (POT) approach is used to build the model, which is then applied to specific reanalysis time series and NOAA buoy records. The model considers the annual and semiannual cycles which are parameterized in terms of harmonic functions. The inclusion of seasonal variabilities substantially reduces the residuals of the fitted model. The information obtained in this study can be useful to design maritime works, because (a) the model improves the understanding of the variability of extreme wave climate along a year and (b) the model accounts for the duration of the storm, which is a key parameter in several formulations for rubble mound breakwater design.     

Optimized support vector regression algorithm-based modeling of ship dynamics

This study contributes to solving the problem of how to derive a simplistic model feasible for describing dynamics of different types of ships for maneuvering simulation employed to study maritime traffic and furthermore to provide ship models for simulation-based engineering test-beds. The problem is first addressed with the modification and simplification of a complicated and nonlinearly coupling vectorial representation in 6 degrees of freedom (DOF) to a 3 DOF model in a simple form for simultaneously capturing surge motions and steering motions based on several pieces of reasonable assumptions. The created simple dynamic model is aiming to be useful for different types of ships only with minor modifications on the experiment setup. Another issue concerning the proposed problem is the estimation of parameters in the model through a suitable technique, which is investigated by using the system identification in combination with full-scale ship trail tests, e.g., standard zigzag maneuvers. To improve the global optimization ability of support vector regression algorithm (SVR) based identification method, the artificial bee colony algorithm (ABC) presenting superior optimization performance with the advantage of few control parameters is used to optimize and assign the particular settings for structural parameters of SVR. Afterward, the simulation study on identifying a simplified dynamic model for a large container ship verifies the effectiveness of the optimized identification method at the same time inspires special considerations on further simplification of the initially simplified dynamic model. Finally, the further simplified dynamic model is validated through not only the simulation study on a container ship but also the experimental study on an unmanned surface vessel so-called I-Nav-II vessel. Either simulation study results or experimental study results demonstrate a valid model in a simple form for describing the dynamics of different types’ ships and also validate the performance of the proposed parameter estimation method.     

Machine learning based mapping of the wave attenuation mechanism of an inclined thin plate

The surface piercing and floating coastal defense structures can be applied as an alternative to conventional rubble mound structures in some specific circumstances. A partially submerged steeply inclined thin plate (ITP) is also one of the candidate alternative structures. Knowledge about the wave attenuation mechanism of ITP improves the engineer's ability to make more cost-effective design. From this motivation, the mechanism of ITP was modeled by artificial neural networks based on experimental data. It is particularly aimed to reveal some fundamental facts about the attenuation mechanism of ITP, which could not be previously attained solely by the conventional analysis of the relevant experimental data. Surface plots, which depict the relationships between the governing design variables were generated from the developed model. In this way, the influence of each individual parameter on the performance was decomposed in a more precise way. Based on the data-driven model outputs, it was inferred that the most dominant design variable is the wavelength. The ITP performance is enhanced with increasing submergence degree, an effect that becomes even more pronounced in severe wave climate conditions. In such wave conditions, decreasing inclination angles also improve the functionality of the structure. However, the generated data-driven model indicated that the combination of the examined variables can have a more complicated effect on the ITP performance, especially for the longer wave lengths.     

Prediction of wave parameters by using fuzzy logic approach

The purpose of this study is to investigate the relationship between wind speed, previous and current wave characteristics. It is expected that such a non-linear relationship includes some uncertainties. A fuzzy inference system employing fuzzy IF–THEN rules has an ability to deal with ill-defined and uncertain systems. Compared with traditional approaches, fuzzy logic is more efficient in linking the multiple inputs to a single output in a non-linear domain. In this paper, a sophisticated intelligent model, based on Takagi–Sugeno (TS) fuzzy modeling principles, was developed to predict the changes in wave characteristics such as significant wave height and zero up-crossing period due to the wind speed. Past measurements of significant wave height values and wind speed variables are used for training the adaptive model and it is then employed to predict the significant wave height amounts for future time intervals such as 1, 3, 6 and 12 h. The verification of the proposed model is achieved through the wave characteristics time series plots and various numerical error criterias. Also the model results were compared with classical Auto Regressive Moving Average with exogenous input (ARMAX) models. For the application of the proposed approach the offshore station located in the Pacific Ocean was used.     

一种利用浮标站资料改进海浪模式有效波高预报的方法

基于浮标站海浪历史数据,利用回归分析方法建立了海浪数值模式有效波高预报产品的一元二次回归方程订正统计模型。通过2017年7月1日-2018年10月10日期间业务试运行结果发现:订正方程能有效改善有效波高数值预报产品的预报精度,且预报时效越短订正效果越显著。其中,第6~11 h预报时效内的订正前后平均绝对误差值减小0.17~0. 241 m,第6~18 h预报时效内订正前后均方根误差减小幅度为0.103~0. 28 m。这说明应用订正统计模型对海浪模式输出产品进行订正,也是改进海浪模式预报准确率的一种有效途径。     

基于支持向量机预测安达曼海南部内孤立波传播特性

内孤立波是发生在密度稳定层化海水中的一种特殊的海洋内波.预测内孤立波传播难度较大.本文提出了一种方法,利用美国麻省理工学院大气环流模型(MITgcm)的内孤立波模型计算了大量模拟数据,建立数据库.采用机器学习的方法,建立一个基于支持向量机(support vector machine,SVM)的安达曼海南部内孤立波传播...