... Estimation with Neural Networks: A Numerical Study from Data of a Spread-Moored FPSO Gustavo A. Bisinotto, Fabio G. Cozman and Eduardo A. Tannuri Escola Polite´cnica, Universidade de Sa o Paulo Sa o Paulo, SP, Brazil ABSTRACT This paper employs neural networks to propose an estimation model for surface...

... residual learning for image recognition," Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778. Hjellbrekke, A-G and Solberg, S (2022) Ozone measurements 2020. NILU, Hu, F, Xia, G-S, Hu, J and Zhang, L (2015). Transferring deep convolutional neural networks for the scene...

... method of neural network-based surrogate model for wave load prediction of offshore platforms Bingquan Yang, Jingxi Liu School of Naval Architecture and Ocean Engineering, Huazhong University of Science and Technology Wuhan, Hubei, China. Hubei Key Laboratory of Naval Architecture & Ocean Engineering...

... Optimization Algorithm, introduces Logistic chaos mapping and adaptive weights (LIWOA) to optimize the weights and thresholds of the neural network, and forms a ship lift duration prediction model based on the LIWOA optimized neural network. The model achieved better prediction results. INTRODUCTION...

... cameras to triangulate the depth for millions of points (pixels). The recent development in deep learning methods for computer vision applications, especially the use of deep convolutional neural networks (CNN) for image recognition (Simonyan 2014), has enabled an acceleration in the development of depth...

... (2012) applied Neural Network (NN) to the estimation of hydrodynamic forces from free running test in a high-speed or medium-speed range. Ouyang, Chen, and Zou (2023) made use of Gaussian process regression (GPR). For a low-speed range, Wakita et al. (2022) used Recurrent Neural Networks (RNN...

... height, which are used to generate various hull forms. To predict the resistance performance of the hull forms scheme, a Back Propagation Neural Network (BP-NN) is constructed based on high-fidelity Computational Fluid Dynamics (CFD) simulation data. The automatic optimization process is formed...

... of them. Notably, the Unet, which is a neural network consisting of consecutive layers of convolution and de-convolution layer, is widely used for this purpose. Meanwhile, numerical simulation is the most time-consuming step of contemporary optimal design for hull form. To avoid the bottleneck caused...

... devised a control method for unmanned vessels that combines Sliding Mode Control and neural networks. Through simulations and sea trials, we confirmed that the amplitude of the unmanned ship's position and heading holding is within the range that allows safe measurement. INTRODUCTION Investigations...

... ABSTRACT This paper proposes a system identification scheme to obtain a multi-input multi-output (MIMO) model of ship motion. The scheme is based on Long Short-Term Memory (LSTM) deep neural network, which can utilize the temporal correlation in the training data of ship maneuvering motion...

... ABSTRACT Among different methods for ship maneuvering motion prediction, efficiency and accuracy are equally important. Considering the ability to identify nonlinear behavior of neural networks, a 4DOF data-driven model is established from the standard mathematical model. Two practical...

... Autonomous Surface Ships (MASS) a reality. As an example, one of the shipping industry's main focuses is implementing such methods to develop digital navigators in autonomous ships. deep learning application neural network machine learning trustworthiness ship perera navigator autonomous ship...

... suction hopper dredger to obtain the motion response parameters of the vessel at different wave angles, periods, and draft depths, which are then compared with experimental data. Based on the deep learning framework, a neural network is constructed, and the obtained vessel motion response data is used...

... are mentioned remote piloting and increasingly autonomous functions on ships (UNCTAD, 2020). artificial intelligence data mining neural network cognition vibration ship algorithm information human perception view machine learning sensor holistic anomaly detection system brain retrieved...

... the detection of debonding. Experiments and electromechanical finite element simulations are conducted. A prediction model of debonding area and shape is established by the improved ELMAN neural network. The time and frequency domain characteristics of the signal are used to train and verify the prediction...

... ABSTRACT In order to improve the accuracy of short-term forecasting of ships, this paper constructs LSTM neural network models with single feature and multi-feature input respectively, studies and predicts the longitudinal motion of ships in waves, analyzes the influence of motion parameters...

... ABSTRACT This study uses deep learning to introduce an automated approach for inspecting a ship's safety plans. We compared Convolutional Neural Network (CNN) models and selected the most suitable for symbol detection. Due to the challenge of obtaining sufficient training data, we proposed...

... Linear Regression and Neural Networks; unsupervised learning, like K-Means Clustering; and reinforcement learning, such as Q-Learning (Sarker, 2021). ML applications span various renewable sectors, including hydro-power turbine business and wind turbine efficiency. In ocean energy, focus has mainly been...

... social responsibility deep learning prediction subsea system sustainability artificial intelligence forecasting mechanism fowt self-attention mechanism offshore wind turbine machine learning platform vector neural network sustainable development dependency sa-lstm semi...

... ABSTRACT In this paper, software-based monitoring system of mooring tension in Floating Offshore Wind Turbine (FOWT) is studied. Long-short-term memory (LSTM) model, which is one of the neural networks (NN), is used as a prediction model to monitor the top tension of the mooring line. Training...