EKC2023

[P23-EE/MO]Prediction of fuel oil consumption for marine diesel engine

Not scheduled

20m

Poster Poster(Wed)

Speaker

Mr Min-Ho Park (Korea maritime and ocean university)

Description

Various studies are being conducted on the prediction of fuel oil consumption of marine diesel engines [1,2]. Fuel oil is a fundamental element that moves ships and accounts for a large portion of operating costs. The vast accumulation of data, the improvement of computer performance, and the development of artificial intelligence technology make it possible to predict marine fuel oil consumption. In this study, fuel oil consumption was predicted based on the data collected from the Hannara, a training ship at Korea Maritime and Ocean University. Data collection was done automatically by HYUNDAI's ACONIS (Advanced Control and Integrated System). The collected data is in MDB (Microsoft Access Database) file and converted to csv to develop data analysis and machine learning model. The data collected included the values of various sensors installed in the marine diesel engine, such as temperature and pressure. The data was divided into training set, validation set, and test set, and the model was trained with the training set, the performance of the model was verified with the validation set, and finally the performance of the model was evaluated with the test set. Fuel oil consumption was predicted with various variables of the collected data, and the performance of the model was evaluated through performance indicators. "This research was supported by the National Research Foundation of Korea (NRF) funded by the Ministry of Education [2022R1F1A10737641260282063490102]."

References

[1] Gkerekos, Christos, Iraklis Lazakis, and Gerasimos Theotokatos. "Machine learning models for predicting ship main engine Fuel Oil Consumption: A comparative study." Ocean Engineering 188 (2019): 106282.
[2] Tran, Tien Anh. "Comparative analysis on the fuel consumption prediction model for bulk carriers from ship launching to current states based on sea trial data and machine learning technique." Journal of Ocean Engineering and Science 6.4 (2021): 317-339.