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.
PO205: A statistical approach to maintenance prioritization in wind turbines using vibration data
Emerson Lima, Manager of Condition Analysis Center, AQTech Power Prognostics
Abstract
Condition monitoring in wind turbines plays a crucial role in early fault detection and in reducing operation and maintenance costs. Despite the widespread use of Condition Monitoring Systems (CMS), the definition of thresholds and alarms still faces significant challenges. Standards such as ISO 20816-21 establish only RMS values in narrow frequency bands (10–2,000 Hz or 10–1,000 Hz), which may not capture typical fault signatures of bearings and gears, often better observed in peak values or in higher frequency bands. Another common practice is to configure alarms at specific kinematic frequencies. In turbines equipped with gearboxes, this may require monitoring around 150 frequencies per machine, resulting in hundreds of alarms that demand considerable effort for configuration and interpretation. Moreover, vibration signals exhibit large variability in magnitude — from micro-g to g — making comparisons difficult. This work presents an alternative methodology based on statistical analysis of equivalent machines within a wind farm. Each vibration trend is reduced to a single indicator by calculating a quantile that represents the baseline. These values, gathered for all machines, are then converted into Z-scores, which express the deviation from the fleet’s average in an adimensional scale. This allows direct comparison across different variables — acceleration, velocity, raw spectrum, or envelope — regardless of units or magnitudes. A key innovation of this approach lies in its data-driven prioritization of maintenance. In large wind farms, especially those operating aging fleets with high fault incidence, maintenance teams face limited resources and cannot address every alarm simultaneously. By ranking machines according to their Z-scores, the methodology enables rapid identification of the most critical assets and early recognition of emerging defects, while deprioritizing less urgent cases. The resulting Z-score matrix functions as a scalable triage tool within the CMS, supporting objective decision-making, reducing alarm complexity, and demonstrating clear applicability across diverse variables.
No recording available for this poster.
