Reducing downtime through predictive or condition‐based maintenance is a promising strategy to help reduce costs associated with wind farm operation and maintenance. To help effectively monitor wind turbine condition, operators now rely on multiply sources of data to make informed operational decisions which can minimise downtime, increasing availability and profitability of any given site. Two of such approaches are SCADA temperature and vibration monitoring, which are typically performed in isolation and compared over time for both fault diagnostics and reliability analysis. Presenting two separate case studies, this paper describes a methodology to bring multiple data sources together to diagnose faults by using a single‐class support vector machine classifier to assess normal behaviour model error, with results showing that anomalies can be detected more consistently when compared to more standard approaches of analysing each data source in isolation.

Anomaly detection for wind turbine condition monitoring is an active area of research within the wind energy operations and maintenance (O & M) community. In this paper three models were compared for multi-megawatt operational wind turbine SCADA data. The models used for comparison were One-Class Support Vector Machine (OCSVM), Isolation Forest (IF), and Elliptical Envelope (EE). Each of these were compared for the same fault, and tested under various different data configurations. IF and EE have not previously been used for fault detection for wind turbines, and OCSVM has not been used for SCADA data. This paper presents a novel method of condition monitoring that only requires two months of data per turbine. These months were separated by a year, the first being healthy and the second unhealthy. The number of anomalies is compared, with a greater number in the unhealthy month being considered correct. It was found that for accuracy IF and OCSVM had similar performances in both training regimes presented. OCSVM performed better for generic training, and IF performed better for specific training. Overall, IF and OCSVM had an average accuracy of 82% for all configurations considered, compared to 77% for EE.

Advancements in wind turbine condition monitoring systems over the last decade have made it possible to optimise operational performance and reduce costs associated with component failure and other unplanned maintenance activities. While much research focuses on providing more automated and accurate fault diagnostics and prognostics in relation to predictive maintenance, efforts to quantify the impact of such strategies have to date been comparatively limited. Through time-based simulation of wind farm operation, this paper quantifies the cost benefits associated with predictive and condition-based maintenance strategies, taking into consideration both direct O&M costs and lost production. Predictive and condition-based strategies have been modelled by adjusting known component failure and repair rates associated with a more reactive approach to maintenance. Results indicate that up to 8% of direct O&M costs can be saved through early intervention along with up to 11% reduction in lost production, assuming 25% of major failures of the generator and gearbox can be diagnosed through advanced monitoring and repaired before major replacement is required. Condition-based approaches can offer further savings compared to predictive strategies by utilising more component life before replacement. However, if weighing up the risk between component failure and replacing a component too early, results suggest that it is more cost effective to intervene earlier if heavy lift vessels can be avoided, even if that means additional major repairs are required over the lifetime of the site.