Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model

Shuo Zhang; Emma Robinson; Malabika Basu

doi:10.1109/UPEC57427.2023.10294560

Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model

Shuo Zhang
, Emma Robinson
, Malabika Basu

School of Electrical and Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Due to the harsh environmental issue and hard accessibility, offshore wind turbines (WTs) have more challenges for operation and maintenance (O&M). Thus, it is crucial to develop effective condition monitoring (CM) methods for WT fault prediction to detect incipient faults before their occurrences, thus preventing durable downtimes. In this paper, eight specific faults are classified for fault prediction using status information from Supervisory Control and Data Acquisition (SCADA) data. The classification steps are based on fault prediction from 10 to 210 minutes prior to faults. By embedding a model-agnostic vector representation for time, Time2Vec (T2V), into Gated Recurrent Unit (GRU), a novel deep learning neural network model, T2V-GRU, is applied for fault classifications. As a result, T2V-GRU successfully predicts over 84.62% of faults and outperforms its counterpart, vanilla GRU, in both overall and individual fault predictions in terms of accuracy, recall scores and F-scores.

Original language	English
Title of host publication	58th International Universities Power Engineering Conference, UPEC 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350316834
DOIs	https://doi.org/10.1109/UPEC57427.2023.10294560
Publication status	Published - 2023
Event	58th International Universities Power Engineering Conference, UPEC 2023 - Dublin, Ireland Duration: 30 Aug 2023 → 1 Sep 2023

Publication series

Name	58th International Universities Power Engineering Conference, UPEC 2023

Conference

Conference	58th International Universities Power Engineering Conference, UPEC 2023
Country/Territory	Ireland
City	Dublin
Period	30/08/23 → 1/09/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Condition monitoring (CM)
Gated Recurrent Unit (GRU)
Operation and maintenance (O&M)
Supervisory Control and Data Acquisition (SCADA)
T2V-GRU
Time2Vec (T2V)
Wind turbines (WTs)

Access to Document

10.1109/UPEC57427.2023.10294560

Cite this

Zhang, S., Robinson, E., & Basu, M. (2023). Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model. In 58th International Universities Power Engineering Conference, UPEC 2023 (58th International Universities Power Engineering Conference, UPEC 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/UPEC57427.2023.10294560

@inproceedings{9bde585cfe014d6aa6ac722bca4a97e9,

title = "Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model",

abstract = "Due to the harsh environmental issue and hard accessibility, offshore wind turbines (WTs) have more challenges for operation and maintenance (O\&M). Thus, it is crucial to develop effective condition monitoring (CM) methods for WT fault prediction to detect incipient faults before their occurrences, thus preventing durable downtimes. In this paper, eight specific faults are classified for fault prediction using status information from Supervisory Control and Data Acquisition (SCADA) data. The classification steps are based on fault prediction from 10 to 210 minutes prior to faults. By embedding a model-agnostic vector representation for time, Time2Vec (T2V), into Gated Recurrent Unit (GRU), a novel deep learning neural network model, T2V-GRU, is applied for fault classifications. As a result, T2V-GRU successfully predicts over 84.62\% of faults and outperforms its counterpart, vanilla GRU, in both overall and individual fault predictions in terms of accuracy, recall scores and F-scores.",

keywords = "Condition monitoring (CM), Gated Recurrent Unit (GRU), Operation and maintenance (O\&M), Supervisory Control and Data Acquisition (SCADA), T2V-GRU, Time2Vec (T2V), Wind turbines (WTs)",

author = "Shuo Zhang and Emma Robinson and Malabika Basu",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 58th International Universities Power Engineering Conference, UPEC 2023 ; Conference date: 30-08-2023 Through 01-09-2023",

year = "2023",

doi = "10.1109/UPEC57427.2023.10294560",

language = "English",

series = "58th International Universities Power Engineering Conference, UPEC 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "58th International Universities Power Engineering Conference, UPEC 2023",

address = "United States",

}

Zhang, S, Robinson, E & Basu, M 2023, Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model. in 58th International Universities Power Engineering Conference, UPEC 2023. 58th International Universities Power Engineering Conference, UPEC 2023, Institute of Electrical and Electronics Engineers Inc., 58th International Universities Power Engineering Conference, UPEC 2023, Dublin, Ireland, 30/08/23. https://doi.org/10.1109/UPEC57427.2023.10294560

Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model. / Zhang, Shuo; Robinson, Emma; Basu, Malabika.
58th International Universities Power Engineering Conference, UPEC 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (58th International Universities Power Engineering Conference, UPEC 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model

AU - Zhang, Shuo

AU - Robinson, Emma

AU - Basu, Malabika

PY - 2023

Y1 - 2023

N2 - Due to the harsh environmental issue and hard accessibility, offshore wind turbines (WTs) have more challenges for operation and maintenance (O&M). Thus, it is crucial to develop effective condition monitoring (CM) methods for WT fault prediction to detect incipient faults before their occurrences, thus preventing durable downtimes. In this paper, eight specific faults are classified for fault prediction using status information from Supervisory Control and Data Acquisition (SCADA) data. The classification steps are based on fault prediction from 10 to 210 minutes prior to faults. By embedding a model-agnostic vector representation for time, Time2Vec (T2V), into Gated Recurrent Unit (GRU), a novel deep learning neural network model, T2V-GRU, is applied for fault classifications. As a result, T2V-GRU successfully predicts over 84.62% of faults and outperforms its counterpart, vanilla GRU, in both overall and individual fault predictions in terms of accuracy, recall scores and F-scores.

AB - Due to the harsh environmental issue and hard accessibility, offshore wind turbines (WTs) have more challenges for operation and maintenance (O&M). Thus, it is crucial to develop effective condition monitoring (CM) methods for WT fault prediction to detect incipient faults before their occurrences, thus preventing durable downtimes. In this paper, eight specific faults are classified for fault prediction using status information from Supervisory Control and Data Acquisition (SCADA) data. The classification steps are based on fault prediction from 10 to 210 minutes prior to faults. By embedding a model-agnostic vector representation for time, Time2Vec (T2V), into Gated Recurrent Unit (GRU), a novel deep learning neural network model, T2V-GRU, is applied for fault classifications. As a result, T2V-GRU successfully predicts over 84.62% of faults and outperforms its counterpart, vanilla GRU, in both overall and individual fault predictions in terms of accuracy, recall scores and F-scores.

KW - Condition monitoring (CM)

KW - Gated Recurrent Unit (GRU)

KW - Operation and maintenance (O&M)

KW - Supervisory Control and Data Acquisition (SCADA)

KW - T2V-GRU

KW - Time2Vec (T2V)

KW - Wind turbines (WTs)

UR - https://www.scopus.com/pages/publications/85178136849

U2 - 10.1109/UPEC57427.2023.10294560

DO - 10.1109/UPEC57427.2023.10294560

M3 - Conference contribution

AN - SCOPUS:85178136849

T3 - 58th International Universities Power Engineering Conference, UPEC 2023

BT - 58th International Universities Power Engineering Conference, UPEC 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 58th International Universities Power Engineering Conference, UPEC 2023

Y2 - 30 August 2023 through 1 September 2023

ER -

Wind Turbine Fault Prediction Based On A Novel Gated Recurrent Neural Network Model

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Fingerprint

Cite this