Fault early warning of wind turbine gearbox based on multi‐input support vector regression and improved ant lion optimization

Yang, Yanjun; Liu, Aimin; Xin, Hongwei; Wang, Jianguo

doi:10.1002/we.2604

Cited by 17 publications

(9 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The temperature is an important SCADA-based indicator of incipient fault processes in gearbox components [8,19,20,[22][23]. In the present study, the condition of the gear bearing has been monitored based on two normal operation models of the bearing temperature.…”

Section: Methodsmentioning

confidence: 99%

“…Temperature can be an important indicator of different types of developing machine problems such as mechanical faults which can give rise to excessive friction generating heat. Therefore, a major focus of the proposed SCADAbased condition monitoring approaches is the temperature-based detection of developing faults in the wind turbine subsystems based on models of the turbine's normal operation behaviour in the absence of operational faults [11][12][13][14][15][16][17][18][19][20][21][22]. The present study focusses on the gear bearing temperature as an indicator of developing gearbox faults.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Early Fault Detection with Multi-target Neural Networks

Meyer

2021

Computational Science and Its Applications – ICCSA 2021

View full text Add to dashboard Cite

Wind power is seeing a strong growth around the world. At the same time, shrinking profit margins in the energy markets let wind farm managers explore options for cost reductions in the turbine operation and maintenance. Sensor-based condition monitoring facilitates remote diagnostics of turbine subsystems, enabling faster responses when unforeseen maintenance is required. Condition monitoring with data from the turbines' supervisory control and data acquisition (SCADA) systems was proposed and SCADA-based fault detection and diagnosis approaches introduced based on single-task normal operation models of turbine state variables. As the number of SCADA channels has grown strongly, thousands of independent single-target models are in place today for monitoring a single turbine. Multi-target learning was recently proposed to limit the number of models. This study applied multi-target neural networks to the task of early fault detection in drive-train components. The accuracy and delay of detecting gear bearing faults were compared to state-of-the-art single-target approaches. We found that multi-target multi-layer perceptrons (MLPs) detected faults at least as early and in many cases earlier than single-target MLPs. The multi-target MLPs could detect faults up to several days earlier than the single-target models. This can deliver a significant advantage in the planning and performance of maintenance work. At the same time, the multi-target MLPs achieved the same level of prediction stability.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Early Fault Detection with Multi-target Neural Networks

Meyer

2021

Computational Science and Its Applications – ICCSA 2021

View full text Add to dashboard Cite

show abstract

“…Yang and Zhang (2020) applied four SCADA parameters, including outlet pressure of the lubrication filter, oil temperature, grid power and main bearing rotor side temperature to monitor the gearbox working conditions. Yang et al (2021) and Zeng et al (2020) selected multidimensional SCADA data that were highly correlated with the lubricating oil temperature as the input of the model to predict the gearbox oil temperature as the health index of the gearbox to detect early incipient faults.…”

Section: Application Of Supervisory Control and Data Acquisition Data...mentioning

confidence: 99%

“…Yang et al (2021) and Zeng et al (2020) selected multi-dimensional SCADA data that were highly correlated with the lubricating oil temperature as the input of the model to predict the gearbox oil temperature as the health index of the gearbox to detect early incipient faults.…”

Section: Application Of Supervisory Control and Data Acquisition Data...mentioning

confidence: 99%

Application of SCADA data in wind turbine fault detection – a review

Yuan

2022

View full text Add to dashboard Cite

Purpose With the rapid increase in the number of installed wind turbines (WTs) worldwide, requirements and expenses of maintenance have also increased significantly. The condition monitoring (CM) of WT provides a strong “soft guarantee” for preventive maintenance. The supervisory control and data acquisition (SCADA) system records a huge amount of condition data, which has become an effective means of CM. The main objective of the present study is to summarize the application of SCADA data to fault detection in wind turbines, analyze its advantages and disadvantages and predict the potential of future investigations on the use of SCADA data for fault detection. Design/methodology/approach The authors first review the means of WT CM and summarize the characteristics of CM based on SCADA data. To ensure the quality of SCADA data, data preprocessing methods are analyzed and compared. Then, the failure modes of the key components are discussed and the SCADA data used for fault detection of each component are compared. Moreover, the fault detection methods for WT are classified and a general framework for fault detection is proposed. Finally, the issues in the WT fault detection method based on SCADA data are reviewed. Findings Based on the performed analyses, it is found that although the fault detection accuracy based on SCADA data is relatively poor, it has low capital expenses and low computational cost. More specifically, when there is scarce fault data, the normal SCADA data can be used to detect the fault time. However, the specific fault type cannot be identified in this way. When a large amount of fault data are accumulated in the SCADA system, it can not only detect the occurrence time of the fault but also identify the specific fault type. Originality/value The main contribution of the present study is to summarize the pre-processing methods for SCADA data, the data required for fault detection of key components and the characteristics of the fault detection model. Then we propose a general fault detection framework for wind turbines based on SCADA data, where the maintenance workers can choose the appropriate fault detection method according to different fault detection requirements and data resources. This article is expected to provide guidance for fault detection based on time-series sensor signals and be of interest to researchers, maintenance workers and managers.

show abstract

“…At present, many results have been achieved in the study of fault diagnosis and prediction of wind turbines [15], [16]. Adouni uses multi-layer ANN technology to build a fault detection and identification method for wind turbines with This work is licensed under a Creative Commons Attribution 4.0 License.…”

Section: Introductionmentioning

confidence: 99%

Study of Wind Turbine Fault Diagnosis and Early Warning Based on SCADA Data

Shi¹,

Liu

Gao

2021

IEEE Access

View full text Add to dashboard Cite

Wind turbine fault diagnosis and early warning are important to reduce wind farm operation and maintenance costs and improve power generation efficiency. In this paper, we take the Supervisory Control and Data Acquisition (SCADA) data as the research object and research wind turbine health data purification, fault diagnosis model building, and unit operation status monitoring from a completely datadriven perspective. Firstly, for the problem that Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm cannot identify high-density anomalous data. An anomaly data processing scheme combining a density clustering algorithm and normal power interval estimation is proposed. The accuracy of extracting health data from wind turbines is improved. Secondly, to address the problem that the eXtreme Gradient Boosting (XGBoost) algorithm has more hyperparameters, we propose an optimization scheme based on the Bayesian Optimization Algorithm (BOA) and tree model for feature weight measurement, which improves the efficiency and accuracy of intuitive mapping from SCADA system monitoring data to fault features. Finally, a wind turbine condition monitoring scheme based on the information fusion of multi-characteristic monitoring parameters is designed. The wind turbine condition monitoring scheme proposed in this paper can warn generator system failure 3.67 hours, gearbox system failure 5.

show abstract

Fault early warning of wind turbine gearbox based on multi‐input support vector regression and improved ant lion optimization

Cited by 17 publications

References 53 publications

Early Fault Detection with Multi-target Neural Networks

Early Fault Detection with Multi-target Neural Networks

Application of SCADA data in wind turbine fault detection – a review

Study of Wind Turbine Fault Diagnosis and Early Warning Based on SCADA Data

Contact Info

Product

Resources

About