Fault Exploratory Data Analysis of Real-Time Marine Diesel Engine Data Using R Programming

摘要

The uncertainties in diesel engine parameters often result in an inaccurate model that give rise to poor fault analysis. A typical marine diesel engine model consists of few subsystems such as freshwater cooling system, lubrication oil cooling system, air cooler system and exhaust gas system. Instead of modeling and validating the simulation results with the real marine diesel engine data to determine the possible faults occurrence, this paper describes an exploratory data analysis on the actual data to identify the faults. R programming is used to carry out the analysis such as k-mean clustering by generating the graphical representation of the data from each subsystem. Exploratory data analysis is used together with the fault tree analysis to analyze the results for possible faults. Based on the data analysis, it is found that there are several faults in the fuel oil system, air cooler system and scavenge air system that requires fault corrective actions that may lead to increase shipping cost.