Dependency in State Transitions of Wind Turbines

Inference on model Residuals for State Abstractions

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

© 2017 IEEE. Turbine states and predicting the transition into failure states ahead of time is important in operation and maintenance of wind turbines. This study presents a method to monitor state transitions of a wind turbine based on the online inference on residuals. In a Bayesian framework, the state transitions are based on a hidden variable relevant for the predictor, namely the information of the current state. Given an underlying predictive model based on a student's t-distribution for the samples and a conditional prior on the state transition, it is shown that state transitions can be abstracted from generated data. Two models are presented: 1) assuming independence and 2) assuming dependence between states. In order to select the right models, machine learning is utilized to update hyperparameters on the conditional probabilities. Comparing fixed to learned hyperparameters points out the impact machine learning concepts have on the predictive performance of the presented models. In conclusion, a study on model residuals is performed to highlight the contribution to wind turbine monitoring. The presented algorithm can consistently detect the state transition under various configurations. Comparing to heuristic interpretations of the residuals, both models can qualitatively inform about the time when a state transition occurs.
OriginalsprogEngelsk
TidsskriftI E E E Transactions on Industrial Electronics
Vol/bind64
Udgave nummer6
Sider (fra-til)4836-4845
Antal sider10
ISSN0278-0046
DOI
StatusUdgivet - 2017

Fingeraftryk

Wind turbines
Learning systems
Turbines
Students
Monitoring

Citer dette

@article{f14b673b7f4a421e850366cb88fdf7b2,
title = "Dependency in State Transitions of Wind Turbines: Inference on model Residuals for State Abstractions",
abstract = "{\circledC} 2017 IEEE. Turbine states and predicting the transition into failure states ahead of time is important in operation and maintenance of wind turbines. This study presents a method to monitor state transitions of a wind turbine based on the online inference on residuals. In a Bayesian framework, the state transitions are based on a hidden variable relevant for the predictor, namely the information of the current state. Given an underlying predictive model based on a student's t-distribution for the samples and a conditional prior on the state transition, it is shown that state transitions can be abstracted from generated data. Two models are presented: 1) assuming independence and 2) assuming dependence between states. In order to select the right models, machine learning is utilized to update hyperparameters on the conditional probabilities. Comparing fixed to learned hyperparameters points out the impact machine learning concepts have on the predictive performance of the presented models. In conclusion, a study on model residuals is performed to highlight the contribution to wind turbine monitoring. The presented algorithm can consistently detect the state transition under various configurations. Comparing to heuristic interpretations of the residuals, both models can qualitatively inform about the time when a state transition occurs.",
author = "J{\"u}rgen Herp and Ramezani, {Mohammad Hossein} and {S. Nadimi}, Esmaeil",
year = "2017",
doi = "10.1109/TIE.2017.2674580",
language = "English",
volume = "64",
pages = "4836--4845",
journal = "I E E E Transactions on Industrial Electronics",
issn = "0278-0046",
publisher = "Institute of Electrical and Electronics Engineers",
number = "6",

}

TY - JOUR

T1 - Dependency in State Transitions of Wind Turbines

T2 - Inference on model Residuals for State Abstractions

AU - Herp, Jürgen

AU - Ramezani, Mohammad Hossein

AU - S. Nadimi, Esmaeil

PY - 2017

Y1 - 2017

N2 - © 2017 IEEE. Turbine states and predicting the transition into failure states ahead of time is important in operation and maintenance of wind turbines. This study presents a method to monitor state transitions of a wind turbine based on the online inference on residuals. In a Bayesian framework, the state transitions are based on a hidden variable relevant for the predictor, namely the information of the current state. Given an underlying predictive model based on a student's t-distribution for the samples and a conditional prior on the state transition, it is shown that state transitions can be abstracted from generated data. Two models are presented: 1) assuming independence and 2) assuming dependence between states. In order to select the right models, machine learning is utilized to update hyperparameters on the conditional probabilities. Comparing fixed to learned hyperparameters points out the impact machine learning concepts have on the predictive performance of the presented models. In conclusion, a study on model residuals is performed to highlight the contribution to wind turbine monitoring. The presented algorithm can consistently detect the state transition under various configurations. Comparing to heuristic interpretations of the residuals, both models can qualitatively inform about the time when a state transition occurs.

AB - © 2017 IEEE. Turbine states and predicting the transition into failure states ahead of time is important in operation and maintenance of wind turbines. This study presents a method to monitor state transitions of a wind turbine based on the online inference on residuals. In a Bayesian framework, the state transitions are based on a hidden variable relevant for the predictor, namely the information of the current state. Given an underlying predictive model based on a student's t-distribution for the samples and a conditional prior on the state transition, it is shown that state transitions can be abstracted from generated data. Two models are presented: 1) assuming independence and 2) assuming dependence between states. In order to select the right models, machine learning is utilized to update hyperparameters on the conditional probabilities. Comparing fixed to learned hyperparameters points out the impact machine learning concepts have on the predictive performance of the presented models. In conclusion, a study on model residuals is performed to highlight the contribution to wind turbine monitoring. The presented algorithm can consistently detect the state transition under various configurations. Comparing to heuristic interpretations of the residuals, both models can qualitatively inform about the time when a state transition occurs.

UR - http://www.scopus.com/inward/record.url?scp=85028713025&partnerID=8YFLogxK

U2 - 10.1109/TIE.2017.2674580

DO - 10.1109/TIE.2017.2674580

M3 - Journal article

VL - 64

SP - 4836

EP - 4845

JO - I E E E Transactions on Industrial Electronics

JF - I E E E Transactions on Industrial Electronics

SN - 0278-0046

IS - 6

ER -