Probabilistic Forecasting of Wave Height for Offshore Wind Turbine Maintenance

James Taylor, Jooyoung Jeon

Research output: Contribution to journalArticle

  • 1 Citations

Abstract

Wind power continues to be the fastest growing source of renewable energy. This paper is concerned with the timing of offshore turbine maintenance for a turbine that is no longer functioning. Service vehicle access is limited by the weather, with wave height being the important factor in deciding whether access can be achieved safely. If the vehicle is mobilized, but the wave height then exceeds the safe limit, the journey is wasted. Conversely, if the vehicle is not mobilized, and the wave height then does not exceed the limit, the opportunity to repair the turbine has been wasted. Previous work has based the decision as to whether to mobilize a service vessel on point forecasts for wave height. In this paper, we incorporate probabilistic forecasting to enable rational decision making by the maintenance engineers, and to improve situational awareness regarding risk. We show that, in terms of minimizing expected cost, the decision as to whether to send the service vessel depends on the value of the probability of wave height falling below the safe limit. We produce forecasts of this probability using time series methods specifically designed for generating wave height density forecasts, including ARMA-GARCH models. We evaluate the methods in terms of statistical probability forecast accuracy, as well as monetary impact, and we examine the sensitivity of the results to different values of the costs.
LanguageEnglish
Pages877-890
JournalEuropean Journal of Operational Research
Volume267
Issue number3
Early online date18 Dec 2017
DOIs
StatusPublished - 16 Jun 2018

Fingerprint

Offshore wind turbines
Wind Turbine
Forecasting
Maintenance
Forecast
Service vessels
Turbine
Turbines
Vessel
Exceed
ARMA Model
GARCH Model
Wind Power
Situational Awareness
Renewable Energy
Costs
Weather
Wind power
Repair
Time series

Cite this

Probabilistic Forecasting of Wave Height for Offshore Wind Turbine Maintenance. / Taylor, James; Jeon, Jooyoung.

In: European Journal of Operational Research, Vol. 267, No. 3, 16.06.2018, p. 877-890.

Research output: Contribution to journalArticle

@article{d38ba6e439a247fc942fa75c04dbc651,
title = "Probabilistic Forecasting of Wave Height for Offshore Wind Turbine Maintenance",
abstract = "Wind power continues to be the fastest growing source of renewable energy. This paper is concerned with the timing of offshore turbine maintenance for a turbine that is no longer functioning. Service vehicle access is limited by the weather, with wave height being the important factor in deciding whether access can be achieved safely. If the vehicle is mobilized, but the wave height then exceeds the safe limit, the journey is wasted. Conversely, if the vehicle is not mobilized, and the wave height then does not exceed the limit, the opportunity to repair the turbine has been wasted. Previous work has based the decision as to whether to mobilize a service vessel on point forecasts for wave height. In this paper, we incorporate probabilistic forecasting to enable rational decision making by the maintenance engineers, and to improve situational awareness regarding risk. We show that, in terms of minimizing expected cost, the decision as to whether to send the service vessel depends on the value of the probability of wave height falling below the safe limit. We produce forecasts of this probability using time series methods specifically designed for generating wave height density forecasts, including ARMA-GARCH models. We evaluate the methods in terms of statistical probability forecast accuracy, as well as monetary impact, and we examine the sensitivity of the results to different values of the costs.",
author = "James Taylor and Jooyoung Jeon",
year = "2018",
month = "6",
day = "16",
doi = "10.1016/j.ejor.2017.12.021",
language = "English",
volume = "267",
pages = "877--890",
journal = "European Journal of Operational Research",
issn = "0377-2217",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Probabilistic Forecasting of Wave Height for Offshore Wind Turbine Maintenance

AU - Taylor,James

AU - Jeon,Jooyoung

PY - 2018/6/16

Y1 - 2018/6/16

N2 - Wind power continues to be the fastest growing source of renewable energy. This paper is concerned with the timing of offshore turbine maintenance for a turbine that is no longer functioning. Service vehicle access is limited by the weather, with wave height being the important factor in deciding whether access can be achieved safely. If the vehicle is mobilized, but the wave height then exceeds the safe limit, the journey is wasted. Conversely, if the vehicle is not mobilized, and the wave height then does not exceed the limit, the opportunity to repair the turbine has been wasted. Previous work has based the decision as to whether to mobilize a service vessel on point forecasts for wave height. In this paper, we incorporate probabilistic forecasting to enable rational decision making by the maintenance engineers, and to improve situational awareness regarding risk. We show that, in terms of minimizing expected cost, the decision as to whether to send the service vessel depends on the value of the probability of wave height falling below the safe limit. We produce forecasts of this probability using time series methods specifically designed for generating wave height density forecasts, including ARMA-GARCH models. We evaluate the methods in terms of statistical probability forecast accuracy, as well as monetary impact, and we examine the sensitivity of the results to different values of the costs.

AB - Wind power continues to be the fastest growing source of renewable energy. This paper is concerned with the timing of offshore turbine maintenance for a turbine that is no longer functioning. Service vehicle access is limited by the weather, with wave height being the important factor in deciding whether access can be achieved safely. If the vehicle is mobilized, but the wave height then exceeds the safe limit, the journey is wasted. Conversely, if the vehicle is not mobilized, and the wave height then does not exceed the limit, the opportunity to repair the turbine has been wasted. Previous work has based the decision as to whether to mobilize a service vessel on point forecasts for wave height. In this paper, we incorporate probabilistic forecasting to enable rational decision making by the maintenance engineers, and to improve situational awareness regarding risk. We show that, in terms of minimizing expected cost, the decision as to whether to send the service vessel depends on the value of the probability of wave height falling below the safe limit. We produce forecasts of this probability using time series methods specifically designed for generating wave height density forecasts, including ARMA-GARCH models. We evaluate the methods in terms of statistical probability forecast accuracy, as well as monetary impact, and we examine the sensitivity of the results to different values of the costs.

U2 - 10.1016/j.ejor.2017.12.021

DO - 10.1016/j.ejor.2017.12.021

M3 - Article

VL - 267

SP - 877

EP - 890

JO - European Journal of Operational Research

T2 - European Journal of Operational Research

JF - European Journal of Operational Research

SN - 0377-2217

IS - 3

ER -