Individuals responsible for blade maintenance need to determine the right data model for tracking the lifetime health of all blades
Wind turbine blade health has become not just an exercise in judicious maintenance practices, but also a concern for the reputation of our industry. Many catastrophic failures are seen by the public or caught on video and shared broadly, often leading to negative press and even community pushback on further development.
It is easy for those who are unfamiliar with wind turbine blades to point out the shortcomings in how our industry maintains blades and tracks blade health. A deeper dive, however, reveals interesting limitations as well as opportunities.
In 2020, EPRI performed a survey of over 100 industry blade experts and discovered that there are surprisingly large discrepancies across industry in the realm of blade defect and damage categorization and management, particularly concerning moderate damage. Further, even with examples of the most severe damage, the experts were split nearly evenly on whether to stop and repair or allow the turbine to continue to run with the damage.
Some of the biggest challenges in the transition to more data-driven maintenance approaches include:
- Blade designs are complex, vary across blade models, are constantly evolving, and are largely considered protected intellectual property by turbine manufacturers
- There are no broadly accepted industry standards on categorizing damage or defect findings
- There is insufficient understanding of how quickly damage propagates, and how environmental and operational conditions drive propagation rates
The challenge can seem monumental, but we believe there are some practical steps that we can take as individuals and as an industry to set us in the right direction. The least flashy but perhaps the most important step for individuals responsible for blade maintenance is determining the right data model for tracking the lifetime health of all blades. A data model includes requirements such as the data platform, parameters documented, sampling frequency, statistical presentation, quality and accuracy requirements, standards adhered to, and spatial sampling. Data tracked includes information from quality control during manufacturing, acceptance inspections/reports, commissioning records, inspections reports, repair reports, extreme load events, lightning data, monitoring data, and wind and operational data.
Cross-industry collaboration is vital towards facilitating more effective blade maintenance practices, such as this upcoming EPRI/ESIG blade maintenance workshop. Another example is the IEA Task 43, which has a risk based maintenance working group, tackling questions of how standardizing certain aspects of damage assessment such as terminology and data requirements can open doors to utilizing advanced decision modeling techniques. Or IEA Task 46, developing techniques to model leading edge erosion. We ask you, reader, how will you get involved?
One easy way to help contribute to the aggregation and dissemination of blade knowledge is to take this five question survey about your practices in blade maintenance. The summarized results will be shared broadly!
Contact us
