Posters

Abstract

General summary Wind turbines are getting bigger and taller and, in the meantime, produce more and more data during their lifetime. And with growing nominal power of each WTG, our responsibility grows simultaneously. Being able to make those data speak will be our main challenge for years to come. In addition, wind farm owners are encouraged to reduce wind farm impacts on health and biodiversity. Most of the wind farms must apply specific curtailments about noise, bat, shadow flicker, grid constraints, etc. How can we maximize European green electricity supply while dealing with those regulations by using more detailed data? How can we use those detailed data to correct any wrong turbine behavior in the shortest delay? Method Huge amount of data required a lot of time and people to be analyzed. Our purpose was to build a tool (with the help of BI-tool) to reduce analysis time and focus on solution and optimization. BayWa r.e. has been focusing on digitalization for a long time in order to optimize the production of renewable energy. Our European engineering network has been working on an in-house tool, which we have recently adapted to the French market. With around 85.000 stops per year on our portfolio level for bat curtailment only we couldn't afford to deeply check their origin one by one. It was also hard to define if this stop was in line with the regulation or in permit violation. Therefore, our approach is to implement a BI-tool dashboard to monitor bat curtailment and acknowledge the good behavior of wind turbines. Each 10 minutes data plot is specified according to specific calculation rules. It gives evidence of wind farm compliancy for local Authority and enable engineer to track potential optimization. This internal tool was then shared with foreign colleagues and new features are continuously adding following specific EU market requirement. Other user cases were also developed, such as noise curtailment, wind sector management, shadow flicker, icing, performance monitoring, etc. Results We are now able to track wrong behaviors related to bat curtailment (starting late, stopping too early, active hysteresis, etc.). We quantified those errors and applied dedicative corrective actions faster than in the past. We are also able to identify and specify underperformances, such as caused by rotor blade icing or wrong parameter settings. Alert is raised to the manufacturer, and we work together to solve the issue. Conclusions Renewable industry must come of age and start to use outstanding number of WTG-data to improve performance and reliability of our assets. Learning Objectives The ambition of our presentation is to show that data are still unexploited material in front of us, and must be emphasized to ensure excellence of our industry. Digitalization and Data analysis is helping the renewable energy sector to unlock the full potential of wind energy. This is not about Artificial Intelligence; this is about empowering engineers to find their own independent way of designing creative solutions.