The application of mega complex projects in the heavy industrial sector have been growing globally for the last ten years. Given there is growth of mega complex projects, the failure rate of these types of projects has also increased. According the EY (who have analyzed 500 completed mega projects from the previous five years): “Of the projects analyzed, 60% experienced schedule delays, and 38% had cost overruns.” Based on research performed by McKinsey & Company, the construction industry needs to change to become more productive and as the industry changes it will look very different five to ten years from now. These changes will only be accelerated by the current COVID-19 pandemic. One solution is the application of smaller standard modular plants or trains that can be designed and constructed quicker and more efficiently. A product-based approach will lead to more of a manufacturing style approach to not only improve productivity but to reduce overall lifecycle costs and schedules and improve overall quality and safety. Further, in times of economic uncertainty, it will reduce the business risk for the project as the business can break the project down into “bite-sized pieces”. Companies will need to be innovative in order to be competitive and have a positive return on investment on their future programs and projects. In the current and future economic environment, an innovative way to execute projects is to utilize a product-based approach. This paper will focus on how to develop a standard modular plant using a product-based approach and provide a case study from a small-scale LNG plant.

Numerical modelling of the Arctic ocean dynamics with real-time simulation capability is useful for designing, developing, testing, and validating the performance of Dynamically Positioned (DP) and Autonomous ships/offshore platforms. Advanced simulation technology needs to be developed to predict the expected loads on these systems due to the complex interactions with environmental disturbances. This paper presents models of waves, currents, wind, and ice that comply with the real-time simulation requirements and adequately capture the dynamic characteristics of the most relevant physical processes. A 3D dispersive numerical model is deployed to predict the wave parameters to be utilized to compute the wave loads on a ship with known Response Amplitude Operators (RAO). A uniform current load is then incorporated in a superposition manner by using a combined wave-current field dispersion relation capable of expressing the wavenumber of an interactive wave-current field. The mean and the gust wind components are added to the resultant force components. A multiple regression-based ice model is used to predict the loads caused by an ice field characterized by varied ice thickness, concentration, floe size, drift speed and directions. The stationkeeping performance of a generic DP-controlled ship subjected to environmental disturbances is evaluated for a range of environmental conditions. The proposed models can help design, develop, and evaluate dynamic positioning and autonomous ship controllers’ performance. Another application may be developing a realistic simulation environment to train conventional, DP-controlled and autonomous ship operators.

Residual stresses are internal self-equilibrating stresses that remain in the component even after the removal of external load. The aforementioned stress when superimposed by the operating stresses on the offshore piping, enhance the chances of fracture failure of the components. Thus, it is vital to accurately estimate the residual stresses in topside piping while performing their fitness for service (FFS) evaluation. In the present work, residual stress profiles of girth welded topside sections of P91 pipes piping are estimate using a machine learning approach. The training and testing data for machine learning is acquired from experimental measurements database by Veqter, UK. Twelve different machine learning algorithms, namely, multi-linear regression (MLR), Random Forest (RF), Gaussian process regression (GPR), support vector regression (SVR), Gradient boosting (GB) etc. have been trained and tested. In order to compare the accuracy of the algorithms, four metrics, namely, Root Mean Square Error (RMSE), Estimated Variance Score (EVS), Maximum Absolute Error (AAE), and Coefficient of Determination (R^2) are used. Gradient boosting algorithm gives the best prediction of the residual stress, which is then used to estimate the residual stress for the simulated input parameter space. In the future work authors shall utilize the residual stress predictions from Gradient boosting algorithm to train the Bayesian Network, which can then be used for estimating less conservative through-thickness residual stresses distribution over a wide range of pipe geometries (radius to thickness ratio) and welding parameters (based on heat input). Furthermore, besides topside piping, the proposed approach finds its potential applications in structural integrity assessment of offshore structures, and pressure equipment’s girth welds.

Due to the quality and error of the data itself, historical automatic identification system (AIS) data was insufficient used to predict navigation risk at sea, but it adequately used to train decision-making neural networks. This paper presents a real AIS ship navigation environment with a rule-based and a neural-based decision processes with frame motion and training the decision network using a deep reinforcement learning algorithm. Rule-based decision-making has several applications in the field of adaptive systems, expert systems, and decision support systems, it also including general ship navigation which regulated by the convention on the international regulations for preventing collisions at sea (COLREGs). However, if someone intend to achieve full unmanned ship navigation without any remote control at the open sea, a rule-based decision-making system cannot be implemented alone. With the growing amount of data, complex sea environment, different collision scenarios, the agent-based decision has become an important role in transportation. For ships, combined rule-based and neural-based decision-making is the only option. It has become progressively challenging to satisfy autonomous decision-making development requirements. This study uses deep reinforcement learning to evaluate the performance of decision-making efficiency under different AIS data input shapes. The results show that the decision neural network trained with AIS data has good robustness and a high ability to achieve collision avoidance. Furthermore, using the same methodology, include instructive guidance for processing radar, camera, ENC, etc., respond to different risk perception tasks in different scenarios. It has important implications for fully unmanned navigation.

This paper covers the generic procedure for the creation and usage of a complete system simulation for propulsion systems of ships with focus on complex hybrid systems. Due to the large number of available components and challenging operational profiles, there is no longer a single best solution, especially in specialized shipbuilding. On top of that even minor changes in the propulsion system can have a substantial impact on plant performance and vessel behavior. Thus, the early design stage is crucial. The design procedure usually incorporates the selection and dimensioning of available components and concepts. Once this has been done, the dynamic behavior can be analyzed using the developed system simulation environment. With the knowledge gained the needed alterations in the concept can be identified and implemented in the subsequent design iteration circle. Based on the example of an offshore anchor handling vessel, the integration potential of a hybrid propulsion system is examined and evaluated according to the aforementioned procedure. The applied tool is a system simulation environment developed in the Department of Marine Engineering, consisting of a distributed co-simulation of the existing ship design environment E4 and Matlab ® Simulink ® .

The ocean wave directional spectrum is an important wave characteristic for maritime safety and navigation. Accurate estimation of directional spectra in real-time is a challenge. In this study we aim to reconstruct the directional spectra from ship motions using a deep convolutional encoding-decoding neural network. In-service measurements of ship motions and wave spectra from a WAMOS II wave scanning radar were used to train the neural network. The data was collected from a frigate type ship for a period of two years. We demonstrate that the deep convolutional encoding-decoding neural network is successful in predicting the directional spectra in real-time. At the same time, we conclude that more data is needed for a better prediction performance, including a more complete coverage of operational conditions.

The obstacle avoidance problem of autonomous surface vessels (ASV) has attracted the attention of the marine control research community for long years. Out of safety consideration, it is important for ASV to avoid all kinds of obstacles like shores, cliffs, floaters and other vessels. Developing a heading and path planning strategy for ASV is the main task and the remaining challenge. Traditional obstacle avoidance algorithms lead to too much computing in working environment. The issue of computation cost can be solved by training obstacle avoidance models with reinforcement learning (RL). By using the RL method, the ASV will choose the most efficient action according to the ASV’s experience it learned from the past. In this paper, RL is adopted to design a decision-making agent for obstacle avoidance. To train the obstacle avoidance model under a sparse feedback environment, hierarchical reinforcement learning (HRL) method is applied. Using this algorithm, better obstacle avoidance performance and longer survival time can be achieved. Memory pool modification and target network modification are also used to smooth the training process of the ASV. Simulation results demonstrate that HRL can make the learning process of un-manned ship’s obstacle avoidance smoother and more effective. Also, the living time of ASVs is improved.

Bores propagating in shallow water transform into undular bores and, finally, into trains of solitons. The observed number and height of these undulations, and later discrete solitons, is strongly dependent on the propagation length of the bore. Empirical results show that the final height of the leading soliton in the far-field is twice the initial mean bore height. The complete disintegration of the initial bore into a train of solitons requires very long propagation lengths, but unfortunately these required distances are usually not available in experimental tests or nature. Therefore, the analysis of the bore decomposition for experimental data into solitons is difficult and requires further approaches. Previous studies have shown that by application of the nonlinear Fourier transform based on the Korteweg–de Vries equation (KdV-NFT) to bores and long-period waves propagating in constant depth, the number and height of all solitons can be reliably predicted already based on the initial bore-shaped free surface. Against this background, this study presents the systematic analysis of the leading-soliton amplitudes for non-breaking and breaking bores with different strengths in different water depths in order to validate the KdV-NFT results for non-breaking bores, and to show the limitations of wave breaking on the spectral results. The analytical results are compared with data from experimental tests, numerical simulations and other approaches from literature.

Reduction of drilling costs in the oil and gas industry and the geothermal energy sector is the main driver for major investments in drilling optimization research. The best way to reduce drilling costs is to minimize the overall time needed for drilling a well. This can be accomplished by optimizing the non-productive time during an operation, and through increasing the rate of penetration (ROP) while actively drilling. ROP has already been modeled in the past using empirical correlations. However, nowadays, methods from data science can be applied to the large data sets obtained during drilling operations, both for real-time prediction of drilling performance and for analysis of historical data sets during the evaluation of previous drilling activities. In the current study, data from a geothermal well in the Hanover region in Lower Saxony (Germany) were used to train machine learning models using Random Forest™ regression and Gradient Boosting. Both techniques showed promising results for modeling ROP.

The study on flow around a hydrofoil has been performed using various methods experimentally and numerically. Here we use a purely data-driven model using deep neural networks (DNNs) to reconstruct the flow fields. Its results are also compared with that obtained by traditional methods. The datasets of flow fields around a static hydrofoil obtained from computational fluid dynamics (CFD) are used for training the DNNs and then the trained data-driven model is utilized to make reconstructions and predictions. 9 different physics informed loss functions, which contain the prediction error and the error measuring the violation of the conservation law, are proposed to train the parameters of DNNs. Effects of the proposed loss functions on reconstruction of velocity field and pressure field are analyzed. Lift and drag forces predicted during the training time are also analyzed. However, the data-driven model based on DNNs with the optimum loss function works well in the interpolation but fails in the extrapolation. The reasons causing the errors are discussed.

Wind energy is the most promising clean, renewable energies to the power industry in the world. More and more wind turbine structures equipped with the larger capacity, taller towers, and longer blades were installed at the offshore/onshore wind farms. But these structures face many harsh environmental conditions, and structural damage and foundation scour are continuously accumulated. It could alter the modal parameter and dynamic response and further reduce the safety of structures. It is a significant challenge on how to accurately estimate the structural states if there is structural damage or foundation scour. For addressing these limitations, a One Dimensional Convolutional Neural Network (1D CNN) method is developed to estimate the structural state. After the Fast Fourier Transform of the acceleration signals, these frequency responses are used as the input to train the 1D CNN, while these states are estimated as the output. A simplified spring-beam model is introduced to simulate the pile-soil interaction, and the effects of the damage and scour on natural frequencies are investigated and compared. The effectiveness and robustness of the proposed 1D CNN method have been numerically investigated by several scenarios associated with the wind turbine structure. Results demonstrate that the 1D CNN method can accurately estimate the structural states, even under a noisy environment. Further, the 1D CNN method can identify the location of damage and scour depth with very high accuracy. This approach may be useful in the on-site structural health monitoring in the wind turbine structure.

Being able to give real time on-board advice, without depending on extensive sets of measured data, is the ultimate goal of the digital twin concept. Ideally, the models used in a digital twin only rely on current in-service data, although they have been built using simulated and possibly some measured data. Working with just the 6-DOF motions of a ship, can the local sea state reliably be estimated using the digital twin concept? Does a general model exist to do so, without the need to measure or simulate the particular ship? In this paper, we discuss how simulations of an advancing ship, subjected to various sea states, can be used to estimate the relative wave direction from in-service motion measurements of the corresponding ship. Various types of neural networks are used and evaluated with simulated data and measured data. In order to study the generalization power of the neural networks, a range of ships has been simulated, with varying lengths, drafts and geometries. Neural networks have been trained on selections of the ships in this extended training set and evaluated on the remaining ships. Results show that the developed neural networks give a remarkable performance in simulation data. Furthermore, generalization over geometry is very good, opening the door to train a general model for estimating sea state characteristics. Using the same model for in-service measurements does not perform well enough yet and further research is required. The paper will include discussion on possible causes for this performance gap and some promising ideas for future work.

Subsea intervention operations may expose wellheads to very high structural loads. Traditionally, structural fatigue was not regarded as a challenge for wellheads, and design codes did not include verification with respect to structural loads. However, development in mathematical modeling and numerical techniques as well as monitoring systems have revealed that semi-submersible rigs with marine riser systems and BOP (Blowout Preventer) can cause quite significant loads on permanently installed equipment. Being able to describe a risers movement under the action of external forces is important in order to calculate the effects on the wellhead. This may be used to predict the expected remaining lifetime of the system. In this article, the existing literature on numerical modeling of a riser system is discussed. Different mathematical models of a riser system were approximated numerically by a finite element method. The models are compared and their correctness is discussed. The error bounds for the finite element solutions are derived. Boundary conditions are further developed to improve realism and ensure compliance with existing standards and guidelines. The sensitivity of such changes are assessed and discussed. A discussion on how to use the numerical simulated data to train a machine learning algorithm to estimate well loading is presented and key parameters affecting the cyclic wellhead loads are identified.

This paper proposes a machine learning agent for automatically navigating a vessel in a confined channel subject to environmental conditions. The agent is trained and tested using a Ship Maneuvering Simulator and is responsible for commanding the rudder, so as to keep the vessel inside the channel with minimum distance from the center line, and to reach the final part of the channel with a prescribed thruster rotation level. The algorithm is based on deep reinforcement learning method and uses an efficient state-space representation. The advantage of using reinforcement learning is that it does not require any expert to directly teach the agent how to behave under particular conditions. The novelty of this work is that: it does not require previous knowledge on the vessel dynamic model and the maneuvering scenario; it is robust against fluctuations of environmental forces such as wind and current; it considers discrete actions of rudder commands emulating the pilot actions in a real maneuver. The developed method is convenient for simulations in scenarios or areas that were never navigated before, in which no previous navigation data can be used to train a conventional supervised learning agent. One direct application for this work is the integration with a realistic fast-time maneuvering simulator for new ports or operations. Both training and validation experiments focused on the unsheltered approach channel of the Suape Port, in Brazil; these experiments were run in a SMH-USP maneuvering simulator (real environmental conditions measured on-site were employed in simulations).

At present, over 1500 offshore wind turbines (OWTs) are operating in the UK with a capacity of 5.4GW. Until now, the research has mainly focused on how to minimise the CAPEX, but Operation and Maintenance (O&M) can represent up to 39% of the lifetime costs of an offshore wind farm, mainly due to the assets’ high cost and the harsh environment in which they operate. Focusing on O&M, the HOME Offshore research project ( www.homeoffshore.org ) aims to derive an advanced interpretation of the fault mechanisms through holistic multiphysics modelling of the wind farm. With the present work, an advanced model of dynamics for a single wind turbine is developed, able to identify the couplings between aero-hydro-servo-elastic (AHSE) dynamics and drive train dynamics. The wind turbine mechanical components, modelled using an AHSE dynamic model, are coupled with a detailed representation of a variable-speed direct-drive 5MW permanent magnet synchronous generator (PMSG) and its fully rated voltage source converters (VSCs). Using the developed model for the wind turbine, several case studies are carried out for above and below rated operating conditions. Firstly, the response time histories of wind turbine degrees of freedom (DOFs) are modelled using a full-order coupled analysis. Subsequently, regression analysis is applied in order to correlate DOFs and generated rotor torque (target degree of freedom for the failure mode in analysis), quantifying the level of inherent coupling effects. Finally, the reduced-order multiphysics models for a single offshore wind turbine are derived based on the strength of the correlation coefficients. The accuracy of the proposed reduced-order models is discussed, comparing it against the full-order coupled model in terms of statistical data and spectrum. In terms of statistical results, all the reduced-order models have a good agreement with the full-order results. In terms of spectrum, all the reduced-order models have a good agreement with the full-order results if the frequencies of interest are below 0.75Hz.

Plunging breaking waves that occur in the vicinity of offshore structures can lead to high impulsive slamming loads, which are significant for the structural loading. The occurrence of plunging breaking waves is usually identified based on criteria that are derived from theoretical analyses and experimental studies. Given a large amount of data, detecting plunging breaking waves can be treated as a typical classification problem, which can be solved by a machine learning approach. In this study, logistic regression algorithm is used together with the experimental data from the WaveSlam project to train a classifier for the detection. Three normalized dimensionless features are introduced based on the measured data for the training. A classifier with respect to four wave parameters (i.e. water depth, wave height, crest height and wave period) is then explicitly developed for detecting plunging breaking waves. It is found that the trained classifier has an accuracy of 98.7% and F 1 score of 99.2% for the tested data. Among the three dimensionless parameters, the ratio of wave height to water depth, H/d, is the most decisive factor for the detection of plunging breaking waves.

The main aim of this paper is to perform the validation of the adaptive Gaussian process regression model (AGPRM) developed by the authors for the Stress Intensity Factor (SIF) prediction of a crack propagating in topside piping. For validation purposes, the values of SIF obtained from experiments available in the literature are used. Sixty-six data points (consisting of L , a , c and SIF values obtained by experiments) are used to train the AGPRM, while four independent data sets are used for validation purposes. The experimental validation of the AGPRM also consists of the comparison of the prediction accuracy of AGPRM and Finite Element Method (FEM) relative to the experimentally derived SIF values. Four metrics, namely, Root Mean Square Error (RMSE), Average Absolute Error (AAE), Maximum Absolute Error (MAE), and Coefficient of Determination ( R 2 ), are used to compare the accuracy. A case study illustrating the development and experimental validation of the AGPRM is presented. Results indicate that the prediction accuracy of the AGPRM is comparable with and even higher than that of the FEM, provided the training points of the AGPRM are aptly chosen.

Truncated mooring system is the foundation of hybrid model test of deepwater floating platform with spread mooring system. This paper presents supervised learning method using linear regression model as the learner to generate empirical formulas which can determine the properties of truncated mooring system given the properties of full-depth mooring system. Thousands of completed truncation tasks are used to train the leaner, and then empirical formulas determining the length, axial stiffness and wet weight of the truncated system are established. The formulas are tested using a set of new truncation tasks. The results indicate that the truncated mooring system can be properly predicted by simply using the formulas.

Accurately modelling a self-propelled vessel in a large amplitude seaway with CFD is very expensive and practically out of reach. The expense is due to the very small numerical time-step required for the propeller rotation and the large mesh size. A method for accurately modelling a propeller while reducing computational cost is desirable. This paper describes the first step towards developing a body force propeller model for unsteady conditions. The purpose of this study is to train a semi-empirical algorithm to accurately prescribe the unsteady body force to model the propeller. The MOERI Container Ship propeller is analyzed with RANS CFD. Open water test data is compared to the RANS CFD results of a steady Moving Reference Frame approach. Harmonic surge is applied to a transient rotating mesh model in open water and the behind condition.

Currently, around 1500 offshore wind turbines are operating in the UK, for a total of 5.4GW, with further 3GW under construction, and 13GW consented. Until now, the focus of the research on offshore wind turbines has been mainly on how to minimise the CAPEX, but Operation and maintenance (O&M) can represent up to 39% of the lifetime costs of an offshore wind farm, due mainly to the high cost of the assets and the harsh environment, limiting the access to these assets in a safe mode. The present work is a part of a larger project, called HOME Offshore (www.homeoffshore.org), and it has as aim an advanced interpretation of the fault mechanisms through a holistic multiphysics modelling of the wind farm. The first step (presented here) toward achieving this aim consists of two main tasks: first of all, to identify and rank the most relevant failure modes within a wind farm, identifying the component, its mode of failure, and the relative environmental conditions. Then, to assess (for each failure mode) how the full-order, nonlinear model of dynamics used to represent the dynamics of the wind turbine can be reduced in order, such that is less computationally expensive (and therefore more suitable to be scaled up to represent multiple wind turbines), but still able to capture and represent the relevant dynamics linked with the inception of the chosen failure mode. A methodology to rank the failure modes is presented, followed by an approach to reduce the order of the Aero-Hydro-Servo-Elastic (AHSE) model of dynamics adopted. The results of the proposed reduced-order models are discussed, comparing it against the full-order coupled model, and taking as case study a fixed offshore wind turbine (monopile) in gearbox failure condition.