Nowadays, most of shipboard maintenance activities rely on a time-based approach, with a schedule of operations prescribed on the basis of technicians’ expertise and manufacturers’ specifications. However, with the increasing complexity of vessels’ facilities and systems, the high economic impact of maintenance costs (which can account for nearly 20-30% of operational expenses), the need for high reliability and efficiency, and the widespread use of data gathering and processing infrastructure, a growing interest in condition-based maintenance (CBM) approach is emerged.
In this sense, Machine Learning (ML) has proven to be a powerful tool in approaching CBM tasks as it can handle high dimensional and multivariate data and extract hidden relationships within them in complex and dynamic environments.
Our research examines the feasibility of monitoring the state of Isotta Fraschini Motori marine diesel engines and their components through the analysis of process signals and ML tools. As a case study, an artificial neural network (ANN) regression model is employed for estimating cylinder exhaust gas temperature (EGT) in normal running condition. The selection of signals and tuning of the network are discussed. The model’s performance is validated by comparing predicted data with experimental measurements and simulations. The research work considers a storage system based on HDF5 files, generated by a custom acquisition diagnostic tool.
Bio - Marco Ceglie
Marco Ceglie is a mechanical engineer. He is currently pursuing a PhD in Applied Mechanics at the Polytechnic University of Bari, supervised by Profs. Giuseppe Carbone and Nicola Menga.
His research focuses on the development of diagnostic systems for diesel engines in marine, industrial, and rail traction applications using machine learning and data analysis tools. His work is supported by Isotta Fraschini Motori, a leading provider of diesel systems.
He also works on tribological projects related to the characterization of the peeling mechanism of films and tapes.
Bio - Giuseppe Giannino
Giuseppe is an Electronic and Telecommunication engineer, graduated in 2018 at Polytechnic of Bari. During his studies he had the opportunity to spend lots of time on research projects abroad (between Ireland – Cork Institute of Technologies, where he worked on optical devices for Telco application and Russian Federation – Kazan National Research Technical University, working on antenna design, optical components and data processing).
He worked as an Automation Test Equipment designer and developer and system integrator at Mermec S.p.A. for the IoT Business Unit, where he was involved on connection systems for micro mobility vehicles. Since July 2021, he has been working at the Innovation and Development Centre of Isotta Fraschini Motori (a Fincantieri company) as Hardware and Software integrator engineer for marine and terrestrial assets. Giuseppe’s main technical skills are: IoT systems, communication protocol, embedded software developments, electromagnetic compatibility and data processing.