Detection of Failed Mooring Lines of Spread-Moored FPSO Using Platform Motions and Artificial Neural Networks

Abstract This study investigates the application of an artificial neural network (ANN) for monitoring mooring failures in a spread-moored Floating Production Storage and Offloading (FPSO) vessel based on platform motion data. Synthetically generated statistical data from time-domain simulations of the FPSO were utilized as input features. A two-step ANN framework was proposed. The first-step ANN identifies whether a mooring group is intact or has failed, while the second-step ANN pinpoints the specific failed mooring line within the identified group. For comparison, a single-step ANN approach that directly estimates a failed mooring line using a single ANN model was also implemented. Hyperparameter optimization was performed using Bayesian and random search approaches, and multiple combinations of statistical input variables were evaluated. Results indicate that the mean value of input variables plays a critical role in detecting mooring failures. The proposed two-step ANN approach has higher detection accuracy than the single-step approach.