Uncertainty in the real-time estimation of ship speed through water

Dalheim, Øyvind Øksnes; Steen, Sverre

doi:10.1016/j.oceaneng.2021.109423

Cited by 22 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It should be noted that no approximation is assumed in this formulation and, therefore, it should be validated with the data, if all three variables ( , , ) are available. On the other hand, the measurements for speed-through-water are known to have several problems, as presented by Dalheim and Steen (2021). Thus, it is recommended to use shaft rpm (and not speed-though-water) as the independent variable while creating data-driven regression models to predict the shaft power.…”

Section: Best Practicesmentioning

confidence: 99%

“…If the prescribed distance is too larger or if the ship is sailing in shallow waters, the Doppler shift is calculated using the reflection from the ocean bottom, i.e., the sensor is in ground-tracking mode, and therefore, it would clearly record the ship's speed-over-ground instead of the speed-though-water. Dalheim and Steen (2021) presented a detailed account regarding the uncertainty in the speed-though-water measurements for a ship, commenting that the speed log sensors are considered to be one of the most inaccurate ones onboard the ship.…”

Section: Inherently Faulty and Incorrect Measurementsmentioning

confidence: 99%

See 1 more Smart Citation

Data Processing Framework for Ship Performance Analysis

Gupta,

Kim,

Steen

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The hydrodynamic performance of a sea-going ship can be analysed using the data obtained from the ship. Such data can be gathered from different sources, like onboard recorded in-service data, AIS data, and noon reports. Each of these sources is known to have their inherent problems. The current work gives a brief introduction to these data sources as well as the common problems associated with them, along with some examples. In order to resolve most of these problems, a streamlined semi-automatic data processing framework for fast data processing is developed and presented here. The data processing framework can be used to process the data obtained from any of the above three mentioned sources. The framework incorporates processing steps like interpolating weather hindcast (metocean) data to ship's location in time, deriving additional features, validating data, estimating resistance components, data cleaning, and outlier detection. A brief description of each of the processing steps is provided with examples from existing datasets. The processed data can be further used to analyse the hydrodynamic performance of a ship.

show abstract

Section: Best Practicesmentioning

confidence: 99%

Section: Inherently Faulty and Incorrect Measurementsmentioning

confidence: 99%

Data Processing Framework for Ship Performance Analysis

Gupta,

Kim,

Steen

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Tentou-se emular um cenário de pior caso para cada tipo de erro, cuidando para não se extrapolar a ordem de grandeza da incerteza referente a cada tipo de medida. O trabalho de [Dalheim and Steen 2021] foi utilizado como referência para incerteza nas medidas de velocidade (3,6%), torque (1%) e rotac ¸ão (1%). Já as incertezas das medidas de temperatura (0,01%), pressão (1%) e fluxo de massa (2%) foram baseadas em [Brun and Kurz 1998].…”

Section: Etapa 3 -Inserc ¸ãO De Incertezas No Datasetunclassified

Aplicação de técnicas de mineração em dados sintéticos para manutenção preditiva: um estudo de caso

Schena

Netto

Becker

2022

Anais Do XXXVII Simpósio Brasileiro De Banco De Dados (SBBD 2022)

View full text Add to dashboard Cite

Este artigo apresenta um caso de aplicação de técnicas de mineração de dados em dados sintéticos para manutenção preditiva de um sistema de propulsão naval, com o objetivo de analisar sua aplicabilidade e adequação na construção de modelos preditivos para manutenção. Na primeira etapa, aplicamos técnicas de mineração de dados no dataset original, e levantamos hipóteses sobre os resultados obtidos com dados sintéticos. Na segunda e terceira etapas, respectivamente, testamos as hipóteses levantadas na etapa inicial através da inserção de desbalanceamento de classes e incertezas de medição. Os resultados mostram formas de se enriquecer conjuntos de dados sintéticos para construção de modelos mais parecidos com cenários industriais reais.

show abstract

“…The given ships are equipped with doppler sonar sensors for log speed measurements. The doppler-based log speed measurements may not be accurate enough to carry-out a hydrodynamic analysis, as pointed out by Dalheim and Steen (2021). Thus, it may be useful to include GPS speed measurements in the analysis along with the longitudinal water current speed from the hindcast.…”

Section: Data Validationmentioning

confidence: 99%