ارزیابی استحکام باند لنگر هنگام انجام عملیات دست زدن به لنگر با استفاده از روش شبکه های عصبی مصنوعی

The risk of vessel capsizing is inherent to anchor handling operations (AHOs). Lessons learned from the Bourbon Dolphin accident reveal that the large static heeling angle could not be prevented due to the lack of awareness of the vessel's stability status, which can be improved with the help of a suitable on-board monitoring system. Therefore, an on-board monitoring system is proposed for assessing stability in terms of the static heeling angle. However, a complete mathematical model is not available for estimating a static heeling angle as a function of operational parameters. Therefore, an artificial neural network (ANN)-based functional relationship has been established between the operational parameters and the static heeling angle. Furthermore, a parametric study has been performed to investigate the effect of neural network topology on network performance. The results show that an ANN topology that contains one hidden-layer is efficient enough to predict a static heeling angle. The correlation coefficient between the ANN model predictions and the target values is 0.999. This result shows that the ANN provides an accurate estimate of the static heeling angle as a function of the operational parameters. Therefore, the proposed mathematical model can be used for assessing a vessel's stability during AHOs.