On the application of change detection techniques for the stability monitoring of fishing vessels

This work includes the description of a novel way to tackle the problem of real time stability monitoring. Instead of taking natural roll frequency as the only variable to estimate the stability level of the vessel, a methodology based on the use of signal-based methods with statistical change detection tools is used for detecting changes in the vessel stability and differentiating between safe and non-safe situations.

This methodology also includes a colour coded risk alarm that informs the crew about the current state of the vessel. Moreover, in the proposed work, the behaviour and performance of this method is analysed using roll motion data of a stern trawler generated by a one degree of freedom nonlinear model in irregular beam waves.

In addition, the obtained results are compared to those from the application of a Fast Fourier Transform (fft)- based methodology previously developed by the authors. The performance of this new proposal has been good, both regarding the estimation of the vessel natural roll frequency and the change detection schemes, showing a better performance than the fft-based method.

However, further analysis is needed to validate these results under more wave conditions and sailing situations.