Ship efficiency forecast based on sensors data collection: improving numerical models through data analytics

Andrea Coraddu, Luca Oneto, Francesco Baldi, Davide Anguita

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

12 Citations (Scopus)

Abstract

In this paper authors investigate the problem of predicting the fuel consumption of a vessel in real scenario based on data measured by the onboard automation systems. The goal is achieved by exploiting three different approaches: White, Black and Gray Box Models. White Box Models (WBM) are based on the knowledge of the physical underling processes. Black Box Models (BBMs) build upon statistical inference procedures based on the historical data collection. Author proposal is a Gray Box Model (GBM) which is able to exploit both mechanistic knowledge of the underlying physical principles and available measurements. Results on real world data shows that the BBM is able to remarkably improve a state-of-the-art WBM, while the GBM is able to encapsulate the a-priory knowledge of the WBM into the BBM so to achieve the same performance of the latter but requiring less historical data.

LanguageEnglish
Title of host publicationMTS/IEEE OCEANS 2015 - Genova
Subtitle of host publicationDiscovering Sustainable Ocean Energy for a New World
Place of PublicationPiscataway, NJ
PublisherIEEE
Number of pages10
ISBN (Electronic)9781479987368
DOIs
Publication statusPublished - 17 Sep 2015
EventMTS/IEEE OCEANS 2015 - Genova - Genova, Italy
Duration: 18 May 201521 May 2015

Conference

ConferenceMTS/IEEE OCEANS 2015 - Genova
CountryItaly
CityGenova
Period18/05/1521/05/15

Fingerprint

Numerical models
Ships
sensor
Sensors
ship
forecast
fuel consumption
automation
Fuel consumption
vessel
Automation

Keywords

  • fuel Consumption
  • gray box model
  • machine learning
  • naval propulsion plant
  • ship efficiency

Cite this

Coraddu, A., Oneto, L., Baldi, F., & Anguita, D. (2015). Ship efficiency forecast based on sensors data collection: improving numerical models through data analytics. In MTS/IEEE OCEANS 2015 - Genova: Discovering Sustainable Ocean Energy for a New World [7271412] Piscataway, NJ: IEEE. https://doi.org/10.1109/OCEANS-Genova.2015.7271412
Coraddu, Andrea ; Oneto, Luca ; Baldi, Francesco ; Anguita, Davide. / Ship efficiency forecast based on sensors data collection : improving numerical models through data analytics. MTS/IEEE OCEANS 2015 - Genova: Discovering Sustainable Ocean Energy for a New World. Piscataway, NJ : IEEE, 2015.
@inproceedings{bd89bf3e43584483969cf11ba7a7429f,
title = "Ship efficiency forecast based on sensors data collection: improving numerical models through data analytics",
abstract = "In this paper authors investigate the problem of predicting the fuel consumption of a vessel in real scenario based on data measured by the onboard automation systems. The goal is achieved by exploiting three different approaches: White, Black and Gray Box Models. White Box Models (WBM) are based on the knowledge of the physical underling processes. Black Box Models (BBMs) build upon statistical inference procedures based on the historical data collection. Author proposal is a Gray Box Model (GBM) which is able to exploit both mechanistic knowledge of the underlying physical principles and available measurements. Results on real world data shows that the BBM is able to remarkably improve a state-of-the-art WBM, while the GBM is able to encapsulate the a-priory knowledge of the WBM into the BBM so to achieve the same performance of the latter but requiring less historical data.",
keywords = "fuel Consumption, gray box model, machine learning, naval propulsion plant, ship efficiency",
author = "Andrea Coraddu and Luca Oneto and Francesco Baldi and Davide Anguita",
year = "2015",
month = "9",
day = "17",
doi = "10.1109/OCEANS-Genova.2015.7271412",
language = "English",
booktitle = "MTS/IEEE OCEANS 2015 - Genova",
publisher = "IEEE",

}

Coraddu, A, Oneto, L, Baldi, F & Anguita, D 2015, Ship efficiency forecast based on sensors data collection: improving numerical models through data analytics. in MTS/IEEE OCEANS 2015 - Genova: Discovering Sustainable Ocean Energy for a New World., 7271412, IEEE, Piscataway, NJ, MTS/IEEE OCEANS 2015 - Genova, Genova, Italy, 18/05/15. https://doi.org/10.1109/OCEANS-Genova.2015.7271412

Ship efficiency forecast based on sensors data collection : improving numerical models through data analytics. / Coraddu, Andrea; Oneto, Luca; Baldi, Francesco; Anguita, Davide.

MTS/IEEE OCEANS 2015 - Genova: Discovering Sustainable Ocean Energy for a New World. Piscataway, NJ : IEEE, 2015. 7271412.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Ship efficiency forecast based on sensors data collection

T2 - improving numerical models through data analytics

AU - Coraddu, Andrea

AU - Oneto, Luca

AU - Baldi, Francesco

AU - Anguita, Davide

PY - 2015/9/17

Y1 - 2015/9/17

N2 - In this paper authors investigate the problem of predicting the fuel consumption of a vessel in real scenario based on data measured by the onboard automation systems. The goal is achieved by exploiting three different approaches: White, Black and Gray Box Models. White Box Models (WBM) are based on the knowledge of the physical underling processes. Black Box Models (BBMs) build upon statistical inference procedures based on the historical data collection. Author proposal is a Gray Box Model (GBM) which is able to exploit both mechanistic knowledge of the underlying physical principles and available measurements. Results on real world data shows that the BBM is able to remarkably improve a state-of-the-art WBM, while the GBM is able to encapsulate the a-priory knowledge of the WBM into the BBM so to achieve the same performance of the latter but requiring less historical data.

AB - In this paper authors investigate the problem of predicting the fuel consumption of a vessel in real scenario based on data measured by the onboard automation systems. The goal is achieved by exploiting three different approaches: White, Black and Gray Box Models. White Box Models (WBM) are based on the knowledge of the physical underling processes. Black Box Models (BBMs) build upon statistical inference procedures based on the historical data collection. Author proposal is a Gray Box Model (GBM) which is able to exploit both mechanistic knowledge of the underlying physical principles and available measurements. Results on real world data shows that the BBM is able to remarkably improve a state-of-the-art WBM, while the GBM is able to encapsulate the a-priory knowledge of the WBM into the BBM so to achieve the same performance of the latter but requiring less historical data.

KW - fuel Consumption

KW - gray box model

KW - machine learning

KW - naval propulsion plant

KW - ship efficiency

UR - http://www.scopus.com/inward/record.url?scp=84957630572&partnerID=8YFLogxK

U2 - 10.1109/OCEANS-Genova.2015.7271412

DO - 10.1109/OCEANS-Genova.2015.7271412

M3 - Conference contribution book

BT - MTS/IEEE OCEANS 2015 - Genova

PB - IEEE

CY - Piscataway, NJ

ER -

Coraddu A, Oneto L, Baldi F, Anguita D. Ship efficiency forecast based on sensors data collection: improving numerical models through data analytics. In MTS/IEEE OCEANS 2015 - Genova: Discovering Sustainable Ocean Energy for a New World. Piscataway, NJ: IEEE. 2015. 7271412 https://doi.org/10.1109/OCEANS-Genova.2015.7271412