Investigation of data preprocessing techniques for ship performance analysis

Ιliopoulos, Panagiotis - Georgios; Ηλιόπουλος, Παναγιώτης - Γεώργιος

dc.contributor.author	Ιliopoulos, Panagiotis - Georgios	en
dc.contributor.author	Ηλιόπουλος, Παναγιώτης - Γεώργιος	el
dc.date.accessioned	2023-03-10T09:37:38Z
dc.date.available	2023-03-10T09:37:38Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/57226
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.24924
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	Big data analytics	en
dc.subject	Ανάλυση δεδομένων	el
dc.subject	Ship performance monitoring	en
dc.subject	Machine learning	en
dc.subject	Data anomaly detection	en
dc.subject	Energy efficiency Emission control	en
dc.subject	Ποσοτικοποίηση της απόδοσης ενός πλοίου	el
dc.subject	Μηχανική μάθηση	el
dc.subject	Ανίχνευση επισφαλών μετρήσεων	el
dc.subject	Έλεγχος εκπομπών ενεργειακής απόδοσης	el
dc.title	Investigation of data preprocessing techniques for ship performance analysis	en
dc.title	Έρευνα τεχνικών προεπεξεργασίας δεδομένων με σκοπό την ποσοτικοποίηση της απόδοσης ενός πλοίου	el
heal.type	bachelorThesis
heal.classification	Data preprocessing	en
heal.classification	Προεπεξεργασία δεδομένων	el
heal.classification	Data preperation	en
heal.classification	Ship performance analysis	en
heal.classification	Ανάλυση της απόδοσης ενός πλοίου	el
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-12-21
heal.abstract	The environmental impact of air emissions produced by the maritime industry is being reduced by increasing the operating energy efficiency of existing ships. An increasing number of vessels are equipped with sensors and devices for monitoring operational behavior, and the amount and access to operational data is gradually increasing. Big data analytics can drastically improve the ship's performance. With the use of proper data preprocessing techniques as well as domain expertise, this research provides an extensive data analytics framework for tracking ship performance under localized operational conditions. A data set from a containership is used to demonstrate the proposed framework. Due to various reasons described in this thesis, the operational data may contain erroneous data points that are critical to assess before performing data analysis or building mathematical and statistical models. The presented investigation relates to detecting data anomalies, identifying the ship's localized operational conditions, calculating the relative correlations among the ship’s operational parameters, quantifying the ship's performance in each of the respective conditions, and the visual representation and analysis of the results. The innovative aspect of this study is the provision of a KPI (i.e., key performance indicator) for ship performance quantification in order to determine the optimal performance trim-draft mode under the engine modes of the case study ship. The suggested framework can be used as an operational energy efficiency measure to provide data quality evaluation and decision support for ship performance monitoring that is valuable to both ship operators and decision-makers.	en
heal.advisorName	Θεμελής, Νικόλαος	el
heal.advisorName	Themelis, Nikolaos	en
heal.committeeMemberName	Σπύρου, Κώστας	el
heal.committeeMemberName	Θεμελής, Νικόλαος	el
heal.committeeMemberName	Παπαδόπουλος, Χρήστος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ναυπηγών Μηχανολόγων Μηχανικών. Τομέας Μελέτης Πλοίου και Θαλάσσιων Μεταφορών	el
heal.academicPublisherID	ntua
heal.numberOfPages	82 σ.	el
heal.fullTextAvailability	false