Robust fault detection approach for wind farms considering missing data tolerance and recovery. Issue 19 (16th February 2021)

• Record Type: Journal Article
• Authors: Zhang, Yuchen; Su, Xiangjing; Meng, Ke; Dong, Zhao Yang
• Is Part Of: IET renewable power generation. Volume 14:Issue 19(2020)
• Page Start: 4150
• Page End: 4158
• Publication Date: 2021-02-16
• Subjects: sensors -- wind power plants -- data acquisition -- fault diagnosis -- condition monitoring -- wind turbines -- offshore installations; robust fault detection approach -- wind farms -- missing data tolerance -- sensing technologies -- real‐time condition monitoring -- wind turbines -- power generation efficiency -- lower the maintenance costs -- real‐time measurements -- Supervisory Control -- Data Acquisition end -- unintended events -- sensor faults -- communication loss -- missing data impact -- robust anomaly detection approach -- WT fault detection -- denoising variational autoencoder -- missing measurements -- high fault detection performance -- realistic offshore WF -- traditional anomaly detection methods -- outstanding robustness -- missing data events -- data recovery; Renewable energy sources -- Periodicals; 333.79405
• DOI: 10.1049/iet-rpg.2020.0604
• Languages: English
• ISSNs: 1752-1416

Abstract

Abstract : The advancement in sensing technologies and infrastructure allows real‐time condition monitoring on wind turbines (WTs), which helps improve the power generation efficiency, lower the maintenance costs of wind farms (WFs). Practically, the real‐time measurements could be unavailable at the Supervisory Control and Data Acquisition end due to unintended events such as sensor faults and communication loss, which significantly depreciates the condition monitoring and fault detection performance. Aiming to mitigate the missing data impact on data‐driven WF applications, this study develops a robust anomaly detection approach for WT fault detection using a denoising variational autoencoder. In presence of missing measurements, the proposed approach can not only sustain high fault detection performance but also recover the missing data as an auxiliary function. The proposed approach is tested on a realistic offshore WF and compared with other autoencoder variants and traditional anomaly detection methods. The testing results verify the outstanding robustness of the proposed approach against missing data events and demonstrate its great potential in missing data recovery.