Posters
Come meet the poster presenters to ask them questions and discuss their work
We would like to invite you to come and see the posters at our upcoming conference. The posters will showcase a diverse range of research topics, and will give delegates an opportunity to engage with the authors and learn more about their work. Whether you are a seasoned researcher or simply curious about the latest developments in your field, we believe that the posters will offer something of interest to everyone. So please join us at the conference and take advantage of this opportunity to learn and engage with your peers in industry and the academic community.
PO499: Linking Atmospheric Conditions to Failure Events: A Data-Driven Approach for Sustainable Predictive Maintenance
Claudia Sequeira, Professor, University of Algarve
Abstract
The aim of this work is to study the influence of atmospheric conditions on maintenance events in wind turbine components, with the broader aim of enabling more sustainable predictive maintenance strategies. Six years of maintenance historic records (January 2019 until December 2024) were aligned with hourly site-specific weather variables, including air temperature, relative humidity, wind speed, and peak gust. The integration of these datasets produced a clean, time-aligned resource that supports both exploratory and predictive analyses. A transparent labelling scheme, based on engineering thresholds, was applied to categorize maintenance events. To identify typical environmental contexts, the study employed smallK clustering, which revealed distinct weather regimes frequently associated with recorded failures. Two focused analyses were conducted on the gearbox: first, a decision tree model was developed to generate interpretable split rules, demonstrating how specific weather variables are linked to failures; second, association rule mining was applied to categorized weather data, highlighting frequent co-occurrences between certain atmospheric conditions and gearbox events. The results showed clear patterns: elevated wind speeds and peak gusts were consistently associated with increased gearbox-related interventions, while periods of high relative humidity correlated with specific failure types. The decision tree analysis provided simple and transparent rules that operators can use directly, while the association rules offered a complementary perspective on conditional probabilities of failures under particular weather regimes. In summary, this study demonstrates that coupling historical maintenance data with contextual weather information not only enhances understanding of external drivers of failures but also provides actionable insights for predictive maintenance planning. By identifying interpretable patterns, the approach contributes to reducing downtime, optimizing maintenance scheduling, and supporting more sustainable asset management in wind energy operations.
No recording available for this poster.
