Analysis of time series imaging approaches for the application of fault classification of marine systems

Velasco, Christian and Lazakis, Iraklis; Leva, Maria Chiara and Patelli, Edoardo and Podofillini, Luca and Wilson, Simon, eds. (2022) Analysis of time series imaging approaches for the application of fault classification of marine systems. In: Proceedings of the 32nd European Safety and Reliability Conference. Research Publishing, IRL. ISBN 9819730000000 (https://doi.org/10.3850/981-973-0000-00-0_esrel202...)

[thumbnail of Velasco-Lazakis-ESREL-2022-Analysis-of-time-series-imaging-approaches-for-the-application-of-fault-classification-of-marine-systems]
Preview
Text. Filename: Velasco_Lazakis_ESREL_2022_Analysis_of_time_series_imaging_approaches_for_the_application_of_fault_classification_of_marine_systems.pdf
Accepted Author Manuscript
License: Strathprints license 1.0

Download (451kB)| Preview

Abstract

Artificial Intelligence (AI) can enable better coordination between ships by enhancing decision-making processes through the optimisation of marine vessels' communication technologies and the gathering of information via Internet of Ships (IoS). Although some efforts have been made to detect faults and malfunctions that can occur in marine systems, there is a lack of analysis and formalisation of fault identification (a.k.a. fault classification) approaches; the aim of which is to provide a comprehensive description of any considered fault type and its respective nature. To contribute to this unexplored field within the shipping sector, an analysis of a total of seven time series imaging approaches (Gramian Summation Angular Field (GASF), Gramian Difference Angular Field (GADF), Markov Transition Field (MTF), Markov Transition Matrix (MTM), Recurrence Plot (RP), compound of GASF-GADF-MTF, and compound of GASF-GADF-MTF-MTM-RP), is performed, as these approaches have demonstrated their ability to identify fault patterns that can not be perceived when considering the original time series data. The resulting images are presented as input in a Convolutional Neural Network (CNN) for the performance of the classification task. As part of this analysis, a case study on the turbocharger exhaust gas outlet temperature parameter of a bulk carrier's main engine is also introduced. Promising results are obtained when the distinct time series imaging approaches are combined, as the compound GASF-GADF-MTF-MTM-RP achieved the maximum accuracy for the analysed case study. Such results evidence the need of exploiting the field of time series imaging for the identification of faults.