A New Intelligent Approach in Predictive Maintenance of Separation System

Abstract

Reducing contaminant emissions is an important task of any industry, included the maritime one. In fact, in April 2018, IMO (International Maritime Organization) adopted an Initial Strategy on reduction of Greenhouse gas (GHG) emissions from ships. An essential part responsible for producing these emissions is the diesel engine. For that reason vessels include separation systems for heavy fuel oils. The purpose of this work is to improve the predictive maintenance techniques incorporating new intelligent approaches. An analysis of vibrations of this separation system was made and their characteristics were used in a Genetic Neuro-Fuzzy System in order to design an intelligent maintenance based on condition monitoring. The achieved results show that the proposed method provides an improvement since it indicates if a maintenance operation is necessary before the schedule one or if it could be possible extend the next maintenance service.

Keywords

• marine fuel separators
• separation system
• predictive maintenance
• greenhouse gas (ghg)
• fast fourier transformation (fft)
• genetic neuro-fuzzy system
• genetic algorithm (ga)
• supervised learning