ORCID

Abstract

Regression-based Machine Learning (ML) approaches are mainly applied to fit the power curve for the performance evaluation of wind turbines (WTs). Although a fitted power curve is prevalent and straightforward for anomaly detection, it is difficult to identify the fault types at the rotor side of a WT, particularly, because the operation can be dependent on multiple parameters. The present paper suggests an interesting approach towards condition monitoring (CM) and fault diagnosis of a DFIG by only processing rotor currents through several signal processing techniques to recognize and localize miscellaneous electrical disturbances. A non-parametric regression approach, Gaussian process regression (GPR), is advised to fit the no-fault performance curve (PC) of rotor current standard deviation (SD) versus wind speed. Thereafter, a hybrid approach with GPR is investigated to visualize no-fault operation, yield the anomaly, and conduct fault recognition at the rotor side with outstanding validation scores in terms of accuracy, dependability, and security.

Keywords

Condition monitoring (CM), Gaussian process regression (GPR), Machine Learning (ML), Performance curve (PC), Standard deviation (SD), Wind turbines (WTs)

Publication Date

2022-01-01

Event

32nd European Safety and Reliability Conference, ESREL 2022

Publication Title

Proceedings of the 32nd European Safety and Reliability Conference, ESREL 2022 - Understanding and Managing Risk and Reliability for a Sustainable Future

Publisher

Research Publishing Services

ISBN

9789811851834

First Page

3127

Last Page

3134

Deposit Date

2026-08-07

Download


Share

COinS