Publications
Detailed Information
A Study on the Method for the Estimation of Energy Efficiency Operational Indicator of a Ship Based on Technologies of Big Data and Deep Learning : 빅데이터와 딥러닝 기술을 기반으로 한 선박 에너지 효율 운항 지표 예측 방법에 대한 연구
Cited 0 time in
Web of Science
Cited 0 time in Scopus
- Authors
- Advisor
- 노명일
- Major
- 공과대학 조선해양공학과
- Issue Date
- 2018-02
- Publisher
- 서울대학교 대학원
- Keywords
- Automatic Identification System (AIS) ; Big data ; Deep learning ; Energy Efficiency Operational Indicator (EEOI) ; Hadoop ; Spark ; Surrogate model
- Description
- 학위논문 (석사)-- 서울대학교 대학원 : 공과대학 조선해양공학과, 2018. 2. 노명일.
- Abstract
- In shipyards, EEOI estimation is required to compare the efficiency of ships and to check the time-varying efficiency of the ships. However, since it is difficult to obtain operating data required for EEOI estimation, it is necessary to estimate EEOI using public data such as Automatic Identification System (AIS) data, ship and engine data, and weather data.
In this study, a method for EEOI estimation using public data was proposed. In the proposed method, total resistance and propeller efficiencies are estimated using the Holtrop-Mennen method, additional resistance is estimated following the International Organization for Standardization (ISO)15016:2015, and engine power is estimated using the modified Direct Power Method (DPM) and Holtrop-Mennen method.
Since the public data have a large capacity, big data technologies such as Hadoop and Spark were applied. The public data was stored to Hadoop, and the data was processed using Spark. Moreover, to reduce the computation time for EEOI estimation, a surrogate model constructed using deep learning was also applied.
To evaluate the effectiveness of the proposed method, it is applied to estimate EEOI of the example ship. The result shows that the method can estimate EEOI effectively and accurately.
- Language
- English
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.