The Offshore Wind Infrastructure Project aims to develop advanced monitoring and testing infrastructure for offshore wind turbines. This includes lifetime component testing, structural health and performance monitoring, and wind measurements. The project seeks to generate data to improve understanding of component lifetime, wind climate conditions, and structural health. It also aims to develop enhanced operations and maintenance strategies. Monitoring systems would acquire vibration, acoustic emission, oil quality, current, temperature, and other data. Damage would be assessed through location, size, and type. Operations and maintenance decision making software would also be developed.

1. The Offshore Wind Infrastructure Project

3. Structural health and Performance Monitoring

4. Wind measurementsGenerate datasets Develop advanced data interpretation and modeling techniques Implement enhanced O&M strategies

5. The Offshore Wind Infrastructure Project Lidar Fix/Floating Climate Chamber Monitoring Systems Acquiring Knowledge of component lifetime Acquiring Knowledge of wind climate in wind farm, wakes, performance Acquiring Knowledge on Structural Health and Performance Monitoring O&M

7. Resonance

8. Fatigue

9. Corrosion

10. Cracks

11. Rotor Blades

12. Surface roughness

13. Icing

14. Fatigue, cracks

15. Loosening at the blade roots

16. Pitch and Yaw System

17. Angle offset

18. Hydraulic failures

19. Motor failure

20. Gearbox CMS Data

21. Offset, misalignment

22. Unbalance

23. Tooth wear

24. Bearing, Shaft

25. Wear, pitting

26. Deformation

27. Fatigue

28. Power Module

29. Failure Frequency convertor

30. Failure Switching Gear

31. Failure Transformer

32. Generator

33. Electrical asymmetries

35. Vibration data

36. Acoustic Emission data

38. Phase currents

40. Temperature

41. Meteorological Data

42. Wind Speed

43. Wind Direction

44. Air Temperature/Pressure

45. Rain

46. Performance Data

47. Power

48. Rotor Speed

49. Blade & Tower SHM Data

50. Vibration data

51. Strain measurements

52. Acoustic Emission data

53. Blade root moments

54. Pitch Yaw Process Data

55. Currents measurements

56. Angles

57. Temperatures

58. Transformer Diagnosis Data

59. Discharge measurements

60. Velocity measurements switches

61. Oil Analysis

62. Temperatures

63. SHM – O&M platform

64. Data Analysis

65. Damage: location, size, type

67. Dynamic Monitoring Challenges How can the dynamic behaviour of a wind turbine be analyzed during operating conditions Existing operational modal analysis techniques are strictly speaking not applicable due to harmonic content of the aerodynamic loads The dynamic behavior of wind turbines is characterized by high aerodynamic damping and nearby modes How to deal with wind turbulences when pitch excitation is used for dynamic testing of operating wind turbines Proposed solutions developed by the VUB/AVRG and to be validated on wind turbines Operational Modal analysis using transmissibility measurements. This is a recently developed OMA techniques that makes no assumption about the nature of the loads The Polyreference Least Squares Frequency domain OMA approach well known for its clear stabilsation diagrams also for highly damped structures and nearby modes OMAX approach combines experimental modal analysis with operational modal analysis by considering both the excitation signal and the turbulence excitation as valuable data Figure from Applicability Limits of Operational Modal Analysis to Operational WindTurbinesD. Tcherniak+, S. Chauhan+, M.H. Hansen Figure from Full-scal modal wind turbine tests: comparing shaker excitation with wind excitation; In Proceedings of IMAC 28

68. Corrosion Monitoring Challenges Can corrosion be predicted for a complex structure like offshore wind turbines located in a harsh environment ? How accurate are these predictions and can the accuracy be improved ? How can the corrosion performance in the splash zone be optimized ? What is the potential for continuously monitoring the state of the CP system / the state of the monopile ? What alternative factors may cause unexpected corrosion. Proposed solutions developed by VUB Lifetime prediction based on a Potential Model, including CP performance. Expanding the capabilities of the model by measuring relevant parameters at the turbine location (O2, salinity, T, pH) + lab-scale experiments using these parameters to generate model input. Screening of potential proven coating alternatives Implementing the SURF corrosion sensor on a test location Focus on biofilms / microbial effects, interior of the pile…

69. Performance Monitoring Challenges Can power curves* obtained from SCADA data be used for fault prediction and diagnosis in wind turbines What is the most adequate approach to estimate and model the power curves of wind turbines Can the detected outliers be correlated with a specific fault Proposed solutions to be investigated Non-parametric modeling based on e.g. data mining approaches Parametric modeling using e.g. Least Squares estimation techniques On-line monitoring by e.g. residual approach, control charts, trend analysis Labeling SCADA data with status/fault codes Figures from On-line monitoring of power curves; Andrew Kusiak*, HaiyangZheng, Zhe Song; Renewable Energy 34 (2009) 1487–1493 *Measurement Power Curve according IEC 61400-12

71. How can the dynamic behavior foundation and support structure be monitored and its design be improved

72. How accurate is prediction about the dynamic behavior of the foundation, tower and blades

73. How can the dynamic behavior wind turbine, tower and blades be monitored during operation

74. Can structural health monitoring be achieved from the measured load data and the identified dynamic behavior

76. Dynamicmonitoring of tower and support structure using operational modal analysis

77. Corrosion monitoring of support structure using modeling and sensors

78. Finite Element Modeling for structural design, load estimation and monitoring

79. PERFORMANCE MONITORING

81. Trend analysis and data miningFinally how can performance monitoring, structural health monitoring and wind data be combined in one improved O&M tool

82. 11 www.bruwind.eu Contact:Dr. ir. ChristofDevriendtVrijeUniversiteitBrussel | Pleinlaan 2 | B-1050 Brussel | BelgiumDept. of Mechanical Engineering | Acoustics & Vibration Research GroupTel. +32 2 6292390 | Fax +32 2 6292865 | GSM +32 477412049Mail: christof.devriendt@vub.ac.be