In water level prediction, Tidal level is a part that satisfies tidal laws, and the residual water level caused by environmental factors is nondeterministic and time-varying. The difficulty of its prediction lies in the inability to quantify short-term meteorological factors such as wind, air pressure, and precipitation, and the increase and decrease of water caused by seasonal climate factors. This is the main reason that restricts the accuracy of traditional water level prediction models. With the development of artificial intelligence, neural networks and fuzzy logic reasoning are gradually applied to marine hydrologic factor prediction. Empirical modal decomposition is a method for handling nonstationary signals that can decompose the different scale fluctuations in the signal step by step. Optimization of residual water level sample data by using empirical modal decomposition method can effectively improve the approximation and prediction ability of the model in machine learning. Based on the above analysis, this paper uses the combined model of Empirical Modal Decomposition and Long Short-Term Memory to separate the data from the residual water level sequence and use machine learning training and prediction to improve the short-term prediction accuracy of the residual water level. Then the predicted residual water level and the harmonic analysis water level are superimposed, that is, the predicted water level. The results show that this method can improve the prediction accuracy of tidal water level to a certain extent and can be used in tidal level prediction. INTRODUCTION Tide is the one of significant dynamic phenomena in about port and harbor activities, and it is also an important factor affecting marine economic activities. With the development of the world economy, trade between countries is becoming more and more frequent. As one of the important transportation modes, maritime shipping occupies a dominant position in the world's transportation industry due to its large capacity and low cost. In recent years, China's shipping industry has made great progress. The leading role of shipping in foreign trade and transportation has become increasingly prominent. The number and types of ships entering and leaving the port have continued to increase. Real-time and accurate water level prediction can provide safety guarantee for large ships entering and leaving ports and diving operations, and also provide a basis for decision-making for shipping establishment. Taking advantage of the strong time correlation of water level changes, accurate water level prediction can reduce the number of tide detection stations and expand the prediction range, thereby saving costs.

In this study, a new integrated methodology (WLRSM) for mapping tidal flats elevation is proposed, which integrates the waterline (WL) method and remotely sensed moisture (RSM) method. Level-1 DEM of tidal flats is provided by WL, and then improved furtherly by RSM. The level-1 DEM was interpolated by waterlines derived from a series of Landsat images in the year of 2019. Meanwhile, the moisture of an exposed tidal flat surface was retrieved using Landsat OLI. We subsequently investigated the relationship between level-1 DEM and moisture, and then constructed statistic models (moisture as the variable) to refine elevations. INTRODUCTION Tidal flats, located in the boundary zone between ocean and land, are sensitive to the interaction between land and sea, and have important ecological functions, coastal protection and social and economic values. The landform of tidal flats is affected by multiple factors, such as sediment transport, hydrodynamic, biological, extreme climate and human activities (Chen, Zhang, Paterson, Thompson, Townend, and Gong, 2017).With the rise of global sea level and the intensification of human activities, tidal flats are facing serious threats. To improve our understanding of the evolution and dynamics of these continually changing ecosystems, frequent measurement of the changing topography of tidal flats is essential. However, owing to rough field conditions, it is difficult to measure the topography of tidal flats with traditional survey techniques. The main methods to study the change of tidal flat morphology include field observation (Gong, Jin, Zhang, Zhou, Zhang, and Li, 2017) and remote sensing retrieval. Field observation (Elgar and Raubenheimer, 2011; Hughes, 2012; Vandenbruwaene, Meire, and Temmerman, 2012) has a high accuracy in obtaining data, but manual measurement, which requires a large amount of manpower and material resources, is required according with the plan, so only topographic data in a limited area and in a limited time can be obtained. Remote sensing (Bell, Bird, and Plater, 2016;Mason, Scott, and Dance, 2010; Ryu, Cho, and Won, 2001) technology can periodically and synchronously monitor a large range of areas and obtain high spatial resolution remote sensing images, which shows great advantages and potential in the study of dynamic evolution of tidal flats (Huang, Chen, Lin, Kuo, Liao, and Lin, 2017; Eom, Jinah, Choi, Jong-Kuk, Ryu, and Joo-Hyung., 2011). Therefore, many methods use remote sensing was proposed to monitor tidal flat topography and try to analyze its evolution trend, such as the waterline (WL) method (Kang, Ding, Xu, Zhang, and Ge, 2017; Mason, Davenport, and Flather, 1997; Mason, Davenport, Flather, and Gurney, 1998) and remotely sensed moisture (RSM) method (Li, Gong, Dai, Lu, Zhang, and Cybele, 2018).

The future knowledge of the waves and force is indispensable for the model identification and the real-time control of ocean engineering devices. In order to effectively control the motion of the offshore structures in a real-time manner, it is required to have an accurate and efficient prediction of the waves. Machine learning has been widely applied in ocean engineering field as it offers compromise between prediction accuracy and computational cost. The present study focuses on wave-force prediction of offshore structures based on deep machine learning algorithms. A novel wave-force prediction model is proposed, which makes full use of the efficient processing characteristics of Long Short-Term Memory Recurrent Neural Network (LSTM RNN) and Nonlinear Autoregressive Exogenous Feedback Neural Network (NARX FNN) for time series data processing. The relationship between the wave height and the wave height is non-causal and nonlinear which need future wave height knowledge for current wave excitation force. Therefore, The LSTM RNN is firstly utilized for multi-step prediction of the time series of wave elevation. The NARX FNN is used to address the model system identification between the wave heights and the wave force. Then, the LSTM RNN is further applied to predict the future force of offshore structures for the real-time control of the structure motions. After that, the proposed deep machine learning algorithm is utilized for wave-force prediction based on the experimental data obtained in Kelvin Hydrodynamic Laboratory and the optimal horizon can be specified for the test model by comparing the performance of different prediction horizons. The results indicate that LSTM-NARX model can successfully predict the time series of the waves and force. 1 INTRODUCTION Wave energy as a dense and stable renewable energy resource is forecast to have the potential to supply 10% of European electricity needs or to generate the equivalent of up to 20% of UK electricity; about half today's total renewable generation. WECs convert the oscillation of kinetic and potential energy carried by ocean gravity waves to electrical energy that can be delivered to the electrical grid through a mechanism known as power take-off system (PTO) (Anderlini,2019) . Wave energy drives two or more parts move relatively then energy is captured by hydraulic mechanic or direct drive. However, there is no wave energy converter reaching commercial stage due to its high levelised cost of energy (LCOE). There is one of the ways to move a step forward by improving the power absorption efficiency under real-time control. By controlling the force exerted from PTO system, such as latching control, it is possible to tune the velocity of WEC with the excitation force of incoming wave for achieving the maximum energy absorption. The wave excitation force is regarded as the combination of incident component and diffraction component from the view of linear potential theory (LPT).

Based on the work by Chen (1993) on the Green function in water of finite depth in which two depth-effect functions are defined, we have performed new analyses and obtained new formulations. The analytical logarithmic terms are extracted from the depth-effect functions. Special numerical algorithms based on integration along complex contours have been developed for numerical computations with an absolute error of order 10 −8 . Furthermore, the formulation is designed that the depth-effect functions are fully smooth varying and then approximated by Chebychev polynomials. Not only the Green's function itself, but also its first-order derivatives and second-order derivatives are numerically evaluated and approximated independently by polynomials. The analytical terms plus polynomials are then used to enhance the effciency and capability of our in-house software HydroStar. Unlike previous studies, the analytical part of the Green's function contains explicitly the logarithmic singularity when the depth-scaled wavenumber tends to zero. Usually included in the "smooth" part of approximation, this singular term is then extracted and the remaining part without singularity is fully smooth and much better approximated with consistent accuracy of O(10 −6 ) and less partitions. Furthermore, the added-mass and damping coeffcients around a hemisphere are given as an example of benchmark results. Finally, the mathematical and physical aspects of the Green's function for wave frequency tending to zero are discussed. INTRODUCTION Being important for hydrodynamic analysis of floating bodies in general and some renewable devices installed in shallow water, in particular, the Green function of wave radiation and diffraction in water of finite depth is re-analysed here. Lots of studies have been done for the Green function of diffraction and radiation in deepwater. As summarized in Xie et al. (2018), there are more than 20 formulations developed for different domains of applications. Unlike the Green function in deep water, however, only few studies on its formulation and effcient computations presented in classical work by John (1950), Endo (1982), Newman (1985, 1992) and Chen (1993). Recently, an increasing interest seems rebounding as shown in the work by Shan et al. (2018), Liu et al. (2018) and Mackay (2019).

The study on viscous-dominated flow-structure interaction (FSI) now trends towards to be popular by using the modern CFD and CSD technique, that concerns in a board range of applications for ship and ocean engineering. For example, CFD and FEA are coupled to predict the dynamic behaviour of a flexible barge in regular head waves, including the hydroelasticity of a large containership and the deformation of a thin flat plate in air flows. The particular phenomenon of interest in this paper involves scattered data interpolation on the interface between the solid and liquid by introduction of the radial basis function (RBF) technology. Additionally, the RBF-based interpolation is optimized and improved according to the region decomposition technology and the error estimate of the RBF, which helps to effectively solve the large full-rank matrix connected with the RBFbased interpolation. In this way, a new data-independent RBF interpolation point selection method is developed. Three typical cases are given by using the global RBF interpolation, the compactly supported RBF interpolation and the partitioned RBF-based interpolation, in which we analyse the bending deformation of the flat plate under the impact of airflow, which is the representative of the FSI problems. By comparison, our developed partitioned RBF-based interpolation shows the global efficiency and accuracy. In particular, the relationship between the error estimate of the RBF and the filling distance of the interpolation domain is well verified. Probably, it provides a new simple approach for data exchange in FSI. INTRODUCTION Destructive problems have arisen in many flow-structure interaction phenomena, such as fatigue failure of propellers, failure of elbow erosion, damage to high-rise buildings by strong winds and bridge deformation due to sea waves. These problems seriously threaten product life and building safety. These problems now are widely solved numerically using the partitioned coupling method due to its efficiency. In the partitioned coupling solution, independent fluid and structural modules can be used respectively. It was verified (Su, 2010; Zhang, 2017) that, fluid-solid modules can help to take advantage of the existing CFD and CSD study within this frame, which usually show different spatial-temporal discrete methods. Thus, it is necessary to develop a high-precision and efficient interface transfer algorithm on the interface between the flow field and the structure field.

Experimental results of sloshing model test have been investigated by data mining in the database of Seoul National University. More than 540 terabytes experimental data have been accumulated for various cargo holds, vessels, environmental conditions, operational conditions, and experimental conditions. The database was organized, cleaned, and analyzed. Model tests related to the floating units larger than standard size LNG carriers or LNG fueled vessels were selected. Then, the selected target data had been used for the artificial neural network to predict the model test results from the test conditions. The network successfully predicted the experimental results considering the various parameters. INTRODUCTION Vessels carry many different liquids such as Liquefied Natural Gas (LNG), petroleum, and sea water in cargo tanks. When a vessel sails under the environmental loads, sloshing occurs in a tank containing partially filled liquid, which can cause serious damage to the structure. Therefore, when designing a ship, it is necessary to accurately estimate the sloshing loads (Faltinsen and Timokha, 2009; Malenica et al., 2017). Recently, many studies have been conducted to estimate sloshing load. One reason is the increase in LNG volume and the rapid expansion of the LNG carrier construction market (BP, 2018; Kuo et al., 2009). LNG carriers have been enlarged for efficient transport. Since larger sloshing impact loads may occur in larger cargo holds, studies have been actively conducted to estimate them. With the expansion of the LNG carrier market, LNG fueled vessels have recently become one of the main products in the shipbuilding market (Congressional Research Service, 2019). Environmental regulations for ship operations have been tightened and LNG has begun to be used as a propulsion fuel for the vessels. Fuel tanks used in the LNG fueled vessels have a different shape from the cargo holds; therefore, partial loading conditions must be considered unlike the LNG carriers.

It is difficult to predict the nominal wake field and optimize the stern shape of the twin-skeg type ship with large breadth and shallow draught due to the huge peak and gradient of the wake field behind the skegs, which is usually the main reason for cavitation deteriorate of the propeller and violent vibration of the stern structure. Besides the model experiment, it is necessary to construct a numerical method with enough accuracy to predict and analyze the wake field of this type ship to meet the requirement of engineering application. Additionally, it is essential to analyze major factors and internal laws influencing wake field of this type ship to improve the uniformity. In this paper, a special CFD numerical method is developed to predict the nominal wake field of a twin-skeg type ro-ro ship with large breadth and shallow draught, in which results of different mesh partition methods and motion freedoms are compared with experiments. On this basis, with displacement and total resistance as constraint conditions, with maximum and mean values of wake field as well as Wake Object Function (WOF) proposed by the Dutch MARIN tank as measurements, influence of skeg shape thinness represented by exit angle and tail radius of 0.8R (R is the propeller radius) water line above the propeller shaft center line is investigated. As results, those factors of skeg have important influence on maximum, mean and WOF values of nominal wake field, which will be the foundation for skeg shape optimization to improve the wake field uniformity. INTRODUCTION Twin-skeg type ship is widely used on ro-ro ships with restricted draft and highly loaded propellers due to the obvious advantage of propulsion efficiency (Kim K et al. 2014). However, it is difficult to design the stern shape of the twin-skeg ship with large breadth and shallow draught due to the huge peak and gradient of the wake field behind skegs, which is one of the main reasons for reduce of propulsion efficiency, deteriorate of cavitation and even excessive vibration of the stern structure. In brief, relationships between stern shape, wake field, propulsion efficiency and cavitation performance are too complicate and internal laws are not enough to support engineering design.

This paper presents numerical modelling results of the interaction between a group of steep waves and a fixed vertical cylinder performed with a one-way coupled hybrid model. A set of experimental data is used to benchmark the accuracy of the modelling results. The wavemaker signal generated in the physical experiments is used to reproduce the incident wave conditions without a priori knowledge of the rest of the dataset. A Lagrangian numerical wave flume propagates the wave group, producing the non-linear free surface elevation and wave kinematics with high accuracy in the vicinity of the cylindrical structure. This set of data is used as the input to the olaFlow CFD model, which calculates the wave-structure interaction on a small computational domain. One-way coupling approaches based on boundary conditions and relaxation zones are tested and compared in terms of the recorded free surface elevation and pressures at the structure. Results present an adequate degree of accordance and turbulence effects are found to be negligible in the simulations. INTRODUCTION Wave and structure interaction (WSI) is a field in which numerical modelling is nowadays being applied consistently and with increasing popularity. The challenges derived from modelling the impact of waves and structures are numerous. Marine and offshore structures are usually located in areas subjected to extremely harsh conditions, in which often waves present three-dimensional and highly-nonlinear processes such as wave breaking and wave impacts may result in impulsive loading. Under such conditions the relevance of numerical modelling, and especially computational fluid dynamics (CFD), derives from the low number of underlying assumptions that the Navier-Stokes equations involve, from their inherent nonlinearity and from their capability to include turbulence dissipation effects via Reynolds-Averaged Navier- Stokes (RANS) or Large Eddy Simulation (LES) turbulence models. Furthermore, despite its advantage and flexibility, numerical modelling results alone have a limited practical credibility if they are not accompanied by a reasonable validation against experimental measurements, which is why physical and numerical modelling should be deemed as complementary approaches.

In harsh sea state, offshore structures may suffer from extreme wave impact loads, which can result in seriously local structural damage. It is still a great challenge in predicting wave impact loads for offshore structures design, since extreme wave impact phenomenon is very complex. In this study, a coupled potential-viscous flow method is applied to analyze focus wave interaction with a fixed vertical cylinder. In the coupled method, an in-house CFD code (Liao et al., 2017; 2018) is coupled with a potential wave generation library MrNWB (Mooring/riser Numerical Wave Basin, Jang H., 2017) based on EOM (Euler Overlay Method, Kim et al., 2012). Wave elevation and impact pressure at typical position are calculated and compared with the benchmark model test data. Results show that the coupled potentialviscous flow method has reasonable good accuracy in predicting wave elevation and wave loads on structure. INTRODUCTION Offshore structures may subject to extreme wave impact, which can result in extreme large impulsive impact loads and even seriously damage of structure. Therefore, it is important to predict wave loads correctly for offshore structure design. However, it is still a great challenge in wave impact loads prediction, since wave impact is a very complex physical phenomenon, involving wave run-up, wave breaking, air entrapped, etc. Great efforts have been made on investigating wave impact phenomenon and predicting wave impact loads on offshore structures. In addition to experimental test (Deng et al., 2016; Wang et al., 2016; Harish et al., 2018), numerical simulation based on CFD is becoming a promising method for wave impact loads prediction but still under development. Numerical investigation of wave run-up on a fixed vertical surfacepiercing cylinder subjected to regular, non-breaking waves using opensource code OpenFOAM has been carried out by Mohammad et al. (2018). Xie et al., (2017) applied a 3D two-phase flow model to study 3D breaking wave interaction with a vertical circular cylinder. Kamath et al., (2016) simulated plunging breaking wave interaction with a vertical cylinder with the open-source CFD code REEF3D. Effect of breaker location was analyzed by changing the location of the cylinder in their work. A RANS based CFD has been used to study focus wave impact on a fixed cylinder by Yoo et al. (2018). Most of the above mentioned research work, a traditional damping zone is commonly applied to absorb refection waves at the outlet boundary. The damping zone should be large enough, usually two times of wave length at least. It results in large computational domain and high computational costs.

ABSTRACT Wave added resistance has great influence on the rapidity and economy of ships, and it is one of the major issues that must be considered in the design of hull lines. In this paper, a panel method based on three dimensional potential flow theory is used to study the problem of wave added resistance. The time-domain motion responses of Wigely-III ship and S175 container ship in head waves are calculated by AQWA, and then the wave added resistances of two ships are obtained by near-field pressure integration method. By comparing the calculated results with the experimental data in literature, it is shown that the variation trend and peak value are in good agreement, and the accuracy and efficiency meet the research requirements. The present method provides an approach of satisfactory accuracy and efficiency to predict wave added resistance of ships voyaging in waves. INTRODUCTION When sailing in waves, the ship will have heave, pitch, roll and other motions, which will cause an increase in resistance comparing with that in calm water. Research shows that the magnitude of wave added resistance could reach 10% ∼ 30% of calm water resistance. With the increasing pressure of energy and environmental protection, IMO has proposed the EEDI formula of the newly built ships, which restricts the energy consumption standard of civil ships more strictly. Therefore, wave added resistance is one of the important factors that should be taken into account in ship's design and optimization stage. Wave added resistance is the component of the steady second-order wave force in the longitudinal direction, which is essentially a nonlinear force. The calculation methods can be generally classified into two categories, namely far-field and near-field methods. The far-field method was firstly introduced by Maruo (1960), and is based on considerations of the diffracted and radiated wave energy and momentum flux at infinity, leading to the steady added resistance force by the total rate of momentum change. The first near-field method was proposed by Boese (1970), which directly integrate the steady second-order pressure acting on the wetted hull surface to obtain wave added resistance. At the initial stage, most of the calculation methods are based on strip theory. Faltinsen et al. (1980) calculated added wave resistance by integrating pressure on average wetted surface of ship hull based on strip method. Arribas (2007) produced a comparison of different methods for the added resistance by using strip theory. With the development of 3D potential flow theory and the improvement of computer performance, 3D time/frequency domain boundary element methods has been widely applied. For example, Liu et al. (2011) employed a 3D frequency domain panel method and a hybrid time domain Rankine source solver based on Green functions and a far-field approach to calculate wave added resistance. Park & Kim (2016) used Rankine source method to solve the motion responses of the ship in head waves, and used the near-field method to obtain wave added resistance, which achieved good prediction accuracy.

ABSTRACT In this study, the proposed linearized inherent strain method for predicting straightening of welded structures is investigated. Thermal elastic-plastic analysis is used to compute inherent strain of a thin flat plate caused by line heating. An accurate gas heat source model is employed in simulating the heating process. Transversal inherent strain is linearized from the computed distortion of the flat plate and applied in the elastic analysis, using higher order Gauss-Legendre (4-nodes shell elements). The accuracy of the proposed method is demonstrated by comparing the angular distortion of both analyses. Furthermore, different plate orientations are taken into account. INTRODUCTION Welding process is one of the most popular assembly method used by many industries such as ships and bridges. However, welding produces many unavoidable problems such as welding residual stress, crack propagation, and welding distortion especially for thin plates and large structures. The welding should be straightened, or the quality of the structure will decrease and the maintenance cost will increase. Welding distortion is caused due to shrinkage during the thermal cycle. It can be separated into the longitudinal shrinkage, transverse shrinkage, and angular distortion, see Fig. 1. In the last decades, ship designers have improved ship performance using thinner and lighter steel plates to reduce the ship weight. However, several problems with welding distortion have appeared as a result of thin plates welding for large structure fabrication. Mechanical and thermal straightening techniques are used to solve this problem. But mechanical techniques are difficult to apply to 3-D structures as ship blocks. Therefore, thermal processes are preferred because it is more economical and more flexible. These techniques create irreversible strain (inherent strain). As discussed by Ueda (1991, 1993), heat straightening consists of applying specific controlled heat patterns to the plastically deformed regions of welded structures in a single or repetitive heating phase and in-plane contraction during the cooling phase. Unfortunately, this method depends most of the cases on the experience of the skilled workers.

ABSTRACT In this study, a FIR filter based displacement estimation technique is proposed to estimate displacement from distributed strain sensors and an accelerometer. First, the neutral axis location of a structure is estimated from strain and acceleration measurements using recursive least square (RLS) method. Then, the low frequency displacement is estimated from the distributed strain gauges and the high frequency displacement from the accelerometer. Finally, these low and high frequency displacements are combined for high fidelity displacement estimation. The feasibility of the proposed method is examined by numerical simulation of a simply-supported beam model under low-frequency and random noise excitations. INTRODUCTION Displacement plays a vital role in structural health monitoring as it provides crucial information regarding structural integrity and its current condition. For example, the displacement response has been employed to identify structural modal parameters of built-up structures (Kim, Kim and Sohn, 2013) and these identified parameters can be used as structural damage indicators. Although the direct measurement of the displacement response is quiet challenging, there are several techniques available. For instance, linear variable differential transformer (LVDT) was applied to measure displacement response of timber railroad bridges (Moreu, Jo, Li, Kim, Cho, Kimmle, Scola, Le, and Spencer, 2013). While the displacement response can be measured with high accuracy and high sampling rate, the installation of LVDT is quite time-consuming. Besides, a fixed support is required as the reference point, which may be impossible in some applications. The global positioning system (GPS) also can be applied for the displacement measurement. This technique was adopted to measure wind-induced displacements of the suspension bridge girder (Nakamura, 2000) and building structures (Breuera, Chmielewski, Górski, and Konopka, 2002; Tamura, Matsui, Pagnini, Ishibashi, and Yoshida, 2002). Although GPS-type technique does not require a fixed reference point, it usually suffers from the low sampling rate and low accuracy problems. Moreover, this technique cannot work under GPS denied environment like underwater conditions.

ABSTRACT Spectral and spatial data from waves are traditionally gathered with single point instruments like buoys or wave gages. These provide accurate timeseries of the sea surface movements but are limited to a singular location. To track the sea surface movement over a larger area is possible with satellite, lidar or radar imaging. The major challenge with these methods is the lack of spatial accuracy. Here optical systems have their merits. This work has the objective to further the principle of stereographic measurement of ocean waves to a practical applicable method. We describe the development of a field employable instrument and practical image recognition algorithms. The results of initial tests and experience with the practical use of the method are presented. Future work will focus on investigation of the accuracy and dependency of the system and to obtain real time analysis. INTRODUCTION The spectral and spatial data from waves are traditionally gathered with single point instruments such as buoys or, wave gages. These provides accurate timeseries of the movement of the sea surface at singular locations. Over the years, significant progress has been made in developing the theory and methods to deduce overall wave dynamics from such singular point data. To track the movement of the sea surface over a larger area can be done with satellite, lidar or radar imaging. The major drawback of these technologies is their lack of spatial accuracy. In this respect, optical systems are regarded to have a performance advantage. Stereographic imaging of ocean waves is described and discussed several publications, some from as early as the beginning of the previous century (Laas, 1905, Mason, 1942). During the last three decades, major advances have been made in digital image acquisition and image processing, making stereographic wave measurements a viable technique today. Recent publications in the field show a progression of the state of art towards practical useable systems. MacHutchon and Liu (2007) demonstrated a system with three cameras. The camera resolution is 640x480 pixels. A so-called area-feature hybrid matching algorithm is used to correlate image pairs to derive a point cloud representing the sea surface. This algorithm contains two steps. First specular reflection is dealt with, to establish matches between corresponding points of the left and right camera images. Secondly, a pixel-based box search is conducted and the matches assessed with normalized cross-correlation. The resulting disparities can then be used to construct a 3d image of the sea plane. Benetazzo (2006) also describes a similar method to derive the 3D sea surface from matched pairs of stereo images. Benetazzo (2006) also uses an algorithm based on finding individual corresponding pixels through cross-correlation. Instead of a regular box search, a so-called pyramidical process (hierarchical) is used, starting with a few pixels on a regular grid on one the camera frames. This grid is refined in subsequent steps. For each pixel of the grid, the corresponding pixels are sought at the equivalent row of the other image. The pixel with the highest cross-correlation coefficient is chosen as the corresponding point, but only when it is above a threshold value. Benatazzo (2006) tested his methodology with both 480*640 resolution cameras and 1008x1008 (1 MP) resolution video cameras. The distance between the cameras was 1.88 m. In subsequent trials, Benetazzo et al. (2012) used a system of two cameras mounted on an oceanographic platform. The distance between the pair of 5 MP cameras was 2.5 m and measuring a 100x100 m area. From the measurements the wave spectra were computed. To calculate the wave spectra, virtual probes were used on in the x-y plane of the surface displacement time series. Benetazzo et al. (2012) found the accuracy to be comparable to traditional wave measurement instruments. Benetazzo et al. (2012) also concluded that the wave spectrum can be derived from the reconstructed spatial sea surface, and that by Fourier transformation of the space time data set, wave dispersion and currents can be estimated. Statistics of crest to through heights further showed good agreement with the Forristall and Tayun-Fedele distributions. Benetazzo et al (2016) discuss the use on moving platforms (vessels) and the necessary calibration procedures. In their publication they emphasize the importance of appropriate systems to transfer the 3D wave data from the camera's image plane to the sea reference plane.

ABSTRACT Discontinuity Layout Optimization (DLO) (Smith and Gilbert, 2007) is a novel geotechnical stability analysis method which is based on the kinematic (upper bound) limit theorem of plasticity. The method is well suited to the development of general-purpose software and is simple to apply in practice, making it an attractive option to competing techniques. Recently, Crumpton (2018) developed an adaptive extension for the DLO method (ADLO) which improves computational efficiency through enhancements to the core solution procedure and the use of automatic mesh refinement. This paper demonstrates the performance of ADLO by applying the method to estimate the bearing capacity of a rough rigid footing on sand overlying clay. INTRODUCTION Discontinuity Layout Optimization (DLO) (Smith and Gilbert, 2007) is a novel geotechnical stability analysis method which is based on the kinematic (upper bound) limit theorem of plasticity. In principle, the method works by identifying an optimum collapse mechanism in a mesh of many potential mechanisms which are constructed by drawing velocity discontinuities between pairs of nodes in a grid. This approach leads to a simple formulation and avoids the numerical complications associated with modelling complete load displacement relationships. It also has a low dependence on a priori information, as meshes do not explicitly employ rigid blocks or finite elements which perform best when oriented according to the stress or velocity field at failure. Recently, Crumpton (2018) developed an Adaptive Discontinuity Layout Optimization (ADLO) algorithm which incorporates several enhancements to the core solution procedure of DLO and a remeshing scheme that uses interpolation error estimates to identify and refine critical regions within the domain of a stability problem. The key benefit of this algorithm is that it requires fewer nodes to attain a given degree of solution accuracy, reducing the amount of computer memory and processing time needed for an analysis. This paper demonstrates the performance of ADLO by applying the method to calculate the bearing capacity of a rough rigid strip footing on sand overlying clay for a range of geometric and material parameters.

ABSTRACT Based on the Reynolds Averaged Navier-Stokes Equations (RANSE) and volume of fluid (VOF) methods of incompressible viscous fluid, the simulation of regular waves is realized under the computational fluid dynamics (CFD) software platform by instantaneously increasing the velocity of the fluid and the wave surface at the fixed wave boundary. Taking the KRISO Container Ship (KCS) as the research object, the numerical simulation study of the ship's resistance in the top wave was carried out. Compared with the model test results, the average error value was 4.10%, and the numerical forecast of the ship's wave increasing resistance in the sea condition of 4–7 was achieved. Based on the numerical simulation method, the irregular waves in the temporary guide of the minimum installed power are simulated, and the minimum installed power is 7967.9 kW through the resistance conversion and self-navigation numerical simulation. INTRODUCTION As global environmental problems get worse, how to reduce the impact of ship operations on the environment is the key to solving the problem. The international maritime organization (IMO) has put forward quantitative requirements on greenhouse gas emissions generated by ship operation and will gradually improve them. Therefore, the ship industry focuses on the research of ship optimization methods to improve the ship's performance. One of the most effective measures to reduce greenhouse gas emissions such as CO 2 is to reduce the main engine power and ship speed, but a smaller power of the main engine may cause the ship's speed loss and cannot maintain its maneuverability. For this reason, IMO has established the minimum propulsion power (IMO, 2013; Zhang, et al., 2016) and other relevant regulations to ensure the safety of the ship under the condition of meeting the ship's energy efficiency design index (EEDI), and the calculation of wave added resistance is the key to calculate the minimum propulsive power.

ABSTRACT The gas kicks occur frequently during the ultra-deepwater high-temperature high-pressure (HPHT) exploration in the South China Sea. They can pose severe risks to well control strategies. The traditional detection method is to analyze the mud-log of dump tanks, which is not only slow but also always having a time-lag between the kick occurrence time and the kick detection time. In this work, a methodology incorporating mud-log and well-log data is proposed for real-time gas kick detection (the time needed to detect kicks is 20 seconds maximum) using artificial intelligence and data-analytics. INTRODUCTION Chen and Ma (2014) proposed that the traditional kick detection method is to analyze the mud-log of dump tanks, which is slow and always having a time-lag between the kick occurrence time and the kick detection time. This is due to the fact that risers are approximately 3000 meters deep. To develop the methodology, mud-log (18 parameters) and well-log data (11 parameters) are collected from 12 kick incidents of 5 drilled wells. Data analysis through artificial intelligence is carried out to develop a kick warning system model considering the impact of semisubmersible platform heave compensation and drilling fluid compressibility. Then, a new comprehensive method incorporating mudlog and well-log data is proposed for real-time kick detection. This is one of the first attempts for real-time kick detection considering mud-log and well-log data using data-analytics and artificial intelligence. This methodology is successfully applied to X gas field in South China Sea and the detection time is only 20 seconds. Technical difficulties of ultra-deepwater HTHP well drilling The South China Sea Basin is controlled by deepwater, high-temperature high-pressure (HPHT). Drilling and completion of deep-water HTHP Well is a new and huge challenge. In the deep-water area of the South China Sea, the water depth of drilled wells has reached 2437.5m, the pressure coefficient is greater than 2.0, and the bottom hole temperature is higher than 200 °C.

ABSTRACT Magnetic detection technology is a special detecting technique developing for submarine detection and location without being affected by light and underwater sound attenuation effect. The aim of this study was to design the reasonable magnetic array with some inductive deep-sea magnetometers to improve the detection efficiency and realize effective location of the submarine pipeline. The Targeting-Location-Theory (TLT) including the magnetic array was constructed by the calculation model. The error analysis was carried out, and the data system and the terminal interface were displayed. It indicated that there is a possibility of developing the magnetic array into real application. INTRODUCTION With the investigation of marine resources and the development process accelerating today, submarine pipeline detection and location has become a necessary part of modern marine undersea engineering. Nowadays, the position information of a lot of submarine pipeline is lost because of various reasons. For instance, the original submarine pipeline doesn't have the map, the map of the pipeline setting is missing, and the previous position of the pipeline is not accurate and so on. Since the diameter of the submarine pipeline is relatively small, the efficient detection and the effective location of the submarine pipeline are going to be two technical focuses. Submarine detection technology development began in the 1960s. Initially relying on divers into the water to detect, but with the increasing depth and the increasing detection range of the submarine object to be detected, relying on artificial direct detection couldn't meet the need. From the beginning of the seventies, Britain and the United States began to develop a variety of submarine pipelines underwater detection equipment (Palshin, 1996). After years of development, it formed three systems including optical detection, acoustic detection and electromagnetic detection in the underwater detection technology. Magnetic detection technology was a special detecting technique developing for submarine detection and location without being affected by light and underwater sound attenuation effect (Caruso, 2000).

ABSTRACT LS25 block in the South China Sea is a typical ultra-deepwater area with HTHP because of its complex geological structure. The block is HTHP, narrow pressure window, it is difficult to accurately predict formation pressure and the drilling cycle is long. Through a new abnormal pressure prediction method of multi-source and multi-mechanism as well as Thermo-Solid coupling calculation model, pre-drilling prediction accuracy has been improved. Wellbore structure is optimized and a safe dual medium continuous circulation drilling system to increasing drilling success rate is developed. Successful drilling in LS25 block is of great significance to the exploration of HTHP. INTRODUCTION Chinese external dependence on oil and gas has exceeded 60%, and the South China Sea is rich in oil and gas resources. The HTHP natural gas resources is up to 15 trillion square meters, and the South China sea is an important replacement area for Chinese energy supply. Zhou (2009) proposed that the South China Sea has long been the focus of competition among countries for a long time Xie et al. (2015) discovered that the exploration and development of oil and gas resources in this region is of great strategic significance to Chinese energy supply and marine rights and interests. Because it is located at the junction of Eurasia, Pacific and Indo-Australia plates, geological structure and abnormal pressure causes are complex (Dong, 2016). In the South China Sea, the deepest water depth reaches 2437.5m, the bottom hole temperature reaches 153 °C and the maximum bottom hole pressure reaches 1.93SG. HTHP drilling and completion in the South China Sea is recognized as a world-class problem. BP, Shell and other international oil companies used the most advanced technology at that time to drill 15 wells in the region. Restricted by the large error of pressure prediction and the complexity of temperature and pressure system, drilling engineering accidents occurred frequently, which cost 4.95 billion yuan but did not achieve the purpose of drilling.

ABSTRACT In the present study, a 168 hrs-forecast wave model is developed to predict significant wave heights at four stations, Japan, using the Group Method of Data Handling (GMDH). The GMDH is one of machine learning (like artificial neural network) and has an advantage of the convenience that once coefficients in a polynomial equation are determined by training, validating and testing data, the forecast can be done by manually solving the polynomial equation. The study uses three different global wave forecast data for inputs produced by JMA, NOAA and ECMWF, and observation data for outputs observed nearshore in Japan. The study found that the combination of ECMWF with others has a superiority to accurately forecast wave information at four stations. INTRODUCTION Waves significantly influence port construction activities that have been empirically determined and governed by a human-based decision that is made by a captain of a vessel or a manager based on a wave condition in a site. The wave height from 0.5 to 1.0 m is a threshold that determines whether the construction can be done or not. Therefore, forecasting 168-hrs wave information is very critical to the decision making to begin the field work. In addition, a high resolution of the wave forecast is one of the critical factors along complex coasts that sufficiently solves wave transformation. Several agencies that use global wave models have forecasted one week ahead-wave information on height, period and direction. In particular, the middle-term forecast has been done by Japan Meteorological Agency (JMA), National Oceanic and Atmospheric Administration (NOAA) and European Center for Medium-Range Weather Forecasts (ECMWF). JMA uses two wave models of Global Wave Model (GWM) and Global Ensemble Model (WEM), NOAA provides 168 hrs-forecast waves using WaveWATCH ΙII (WW3) and ECMWF uses the High RESolution Wave model (WAM). In particular, JMA-GWM provides wave forecast information to 264 hours ahead at every 12 UTC with the 0.5 degree resolution 4 times a day in a global scale ( http://www.jmbsc.or.jp/jp/online/file/f-online20100.html ). NOAA-W W3 forecasts waves to 180 hours ahead with the same resolution and resolution to JMA-GWM ( http://polar.ncep.noaa.gov/waves/implementations.php ). On the other hand, ECMWF-WAM predicts waves to 240 hours ahead with the 0.125 degree resolution 2 times a day ( https://www.ecmwf.int/en/forecasts/datasets/set-ii ). However, those did not guarantee: their accuracies are not sufficiently high to directly apply those to nearshore areas for the decision making in the complex coast in Japan.

ABSTRACT For early recognition of flow regime in pipeline-riser system, use of signals from accelerometers are suggested instead of a commonly used pressure gauge signal. A main factor that discriminates between stable flow and severe slugging is drastically decreased vibration in liquid accumulation of the severe slugging. In this study, a support vector machine is employed for binary classification to identify between stable flow and severe slugging. For multi-class classification, a neural network is applied to recognize four classes of stable flow, two types of severe slugging, and irregular transition. The performance is also analyzed based on the signal length employed. INTRODUCTION Pipeline-riser structures are generally used in offshore oil and gas field. Transported mixtures of liquids, gases, and solid components through horizontal (or inclined) pipeline are raised up to topside via the vertical riser. In the pipeline-riser system, a typical severe slugging can be produced with undesired blowout. It can cause liquid overflow, low production, high pressure in separators, corrosion increase, overload on gas compressors, and extra fatigue induced by cyclic impact (Hill and Wood, 1994; Kang, 1996; Pedersen, 2017; Schmidt, 1985; Sun and Jepson, 1992; Yocum, 1973). The severe slugging should be detected and controlled as quickly as possible to ensure flow assurance for safety and economic benefits. Many studies have classified multiphase flow regimes using machine learning algorithms (Goudinakis, 2004; Mi, 1998; Rosa, 2010; Trafalis, 2005; Wu, 2001; Ye and Guo, 2013; Zou, 2017). For the pipeline-riser system, Goudinakis (2004) investigated an experiment in a S-shaped riser system and measured differential pressures. An input of a neural network (NN) was the normalized differential pressures in 100 seconds. Classes of bubble, oscillation, slug, and severe slugging 1 were identified. Ye and Guo (2013) also performed an experiment in the S-shaped riser to acquire differential pressures. The length of sample was 20 minutes. Least square support vector machines (LS-SVM) were trained with statistical features including mean, variance, skewness, and power spectral densities. Five classes of severe slugging 1, severe slugging 2, severe slugging transition, oscillation, and stable flow were recognized. Zou (2017) obtained differential pressure in the pipeline riser system, calculated feature vectors using mean and range of differential pressures, and adopted the LS-SVM. Samples of various length were analyzed, especially the shortest samples of 6.8 seconds were tested based on the LS-SVM trained by samples of 8.99 seconds. Severe slugging, oscillation, stable like flow, and stable flow were classified. The commonly used pressure gauge has advantages in low cost and application at wide range of temperatures and pressures. However, a pressure tap can be blocked; the pressure can be affected not only by flow regime but also by the phase's velocity (Rosa, 2010); and it is difficult to move into another place after its first installation.