Reducing Predictive Risk in Repowering Energy Assessments through Standardized Operational Data Inputs

Presentation Description: Accurate analysis of the operational history of a wind farm is critically dependent on not only the quality of the operational data recorded by the operator, but also correctly capturing the types of data necessary for predicting energy production of a repowered wind farm. Operational data formats and summaries vary widely between individual plant operators, and may be inconsistent in presentation through the lifetime of the wind farm, particularly in cases of ownership or O&M provider changes.

One of the most important facets of repowering energy assessments is determining the applicability of historical performance factors to going-forward production, as there will be inherent improvements in upgrading to newer or more efficient technology, i.e. improvements in contractual availability definitions due to a change in OEM, site-specific power curve performance, high-wind hysteresis, and other individual loss categories. Clear identification of wind farm production losses fitting these categories allows for more granular assessment of their estimated changes with a repowered wind farm, which in turn leads to higher confidence in both the predicted P50 net energy outputs and uncertainty characterization compared to more simplified production datasets.

This study identifies key production data categories and their benefits to operational repowering assessment accuracy, in an effort to drive improved standardization of recorded data by wind farm operators.