Optimization of spread mooring systems with artificial neural networks

An Artificial Neural Network (ANN) is an information processing technique that is inspired by biological nervous systems. This technique tries to simulate its learning process and uses a mathematical or computational model to process information. Artificial Neural Networks (ANNs) are utilized in many scientific disciplines extensively. There are many problems in mathematics, medicine, business and finance can be solved by using ANNs. The application areas of ANNs are classification, clustering, forecast, function approximation, optimization etc. The prediction of the motion responses for marine systems is computationally very expensive and highly time-consuming. Several methods such as ANNs can be used in order to eliminate high computational costs for these systems. This paper employs ANNs to optimize multiple point mooring systems (spread mooring systems). These systems are often affected from harsh environmental conditions at open seas. To reduce effects of environmental conditions, these systems are supported by mooring lines. In this study, at first OrcaFlex software is employed to model and simulate spread mooring systems. 4 points mooring system is designed in OrcaFlex to calculate mooring tensions and tanker motion displacements. These simulation results are then used to train ANN structure and an algorithm is created. After creating algorithm, the results of training, test and validation are obtained from ANN. Utilization of the ANN tool can provide physical results and can process information in extremely rapid mode promising high accuracy of prediction.