Dynamic stiffness of polyester fiber mooring ropes: Experimental investigation based on radial basis function neural networks

Abstract

Traditional steel wires are being replaced with mooring ropes made from synthetic fibers in offshore industries. However, under the long-term influence of a wave load, the dynamic response of a rope is extremely complex. This paper presents a comprehensive experimental study of the nonlinear dynamic stiffness of polyester fiber mooring ropes under cyclic loading. The stress–strain characteristics of the fiber ropes were investigated based on experimental data. Moreover, cyclic load tests with different mean loads, load amplitudes, and cycles were conducted to study the effects of these three parameters on dynamic stiffness. In previous studies, the dynamic stiffness of fiber ropes was estimated using empirical equations. In this study, the self-learning ability of a radial basis function (RBF) neural network was used to predict the dynamic stiffness of polyester fiber ropes. The predicted results were compared with the empirical formula calculations and experimental measurement results. The RBF neural network has better prediction results than the empirical formulation. In addition, the autonomic learning of the neural network was better when the sample data were in a disordered input state than in an ordered input state. The research findings of this study provide new approaches for solving the dynamic stiffness of ropes and lay a theoretical foundation for the mechanical analysis in the engineering design stage of mooring systems.