Optimizing life cycle cost of wind turbine blades using predictive analytics in effective maintenance planning

A wind turbine blade is capital equipment vital enough to be protected and maintained for inherent safety and reliability during lifetime due to its high impact on turbine availability in event of failure / repair. Unlike matured industries like aerospace, there are no specific guidelines for maintenance plans and mostly the repairs are reactive in nature. This leads to very high cost of maintenance owing to longer downtime of the turbine raising a need to derive an effective maintenance strategy demanding reliability centered maintenance, also facilitating business decisions on spares, service and maintenance requirements through use of available field information, supported by a predictive analytics and reliability models with an overall objective of reducing the operation cost and gaining higher levels of reliability. This paper is an attempt to make use of the widely practiced Predictive Analytics techniques in wind domain to address such challenges and remain competitive in the market. The model built was able to take inputs from different stages of the product life cycle providing a mathematical relationship with respect to failures and contributing factors, allowing addressing the blades that are in critical need of inspection and maintenance at any given point of time based on the rate of wear out. This further becomes a critical input for maintenance planning thereby reducing the operational cost and also attaining high levels of Reliability. Additionally, the model built also provides feedback to the different stages of blade life cycle in terms of setting targets that are required in order to maintain a certain level of Reliability in the field.