Posters

Abstract

When running operations, it is often a struggle to prioritise actions across wind farms because failures are rare, diverse, and inconsistently distributed. A typical 50-turbine site may log only a few hundred failures requiring on-site attention per year, spanning dozens of alarm codes with multiple possible root causes. Classical statistical approaches provide limited value in this context, and fragmented records make it difficult to understand how operational factors interact with technical issues. A key barrier to optimising operations is the availability and quality of operational data. Without a clear and reliable event history, it is challenging to determine which issues have the greatest impact on overall performance, or whether losses are influenced more by turbine hardware, troubleshooting quality, spare part availability, logistics, or operational processes. We will present our work on building data foundations for performance tracking on operational sites that combine automated event classification with empirical knowledge and technical domain expertise. This approach is not a silver bullet solution but a structured process: establishing a transparent baseline, integrating data meaningfully, selecting context-appropriate metrics, and training teams to interpret them. Operators can use these metrics to understand how multiple factors interact and prioritise actions more effectively. By connecting technical insights to operational processes, the framework provides clarity on the causes of underperformance and helps operators make more informed decisions. We will show practical approaches to data management, digitalisation in O&M, and operational strategies, showing how raw operational data can be transformed into actionable insight to support better evaluation of performance and improvement initiatives across wind farms.