SCADA data

SCADA stands for supervisory control and data acquisition, and it is the system that runs and monitors a wind farm. Every modern turbine reports a continuous stream through it: power output, wind speed and direction, rotor and generator speed, pitch and yaw positions, component temperatures, and fault codes. The convention across the industry is ten-minute averages, often with minimum, maximum, and standard deviation alongside the mean.

The ten-minute interval is a historical compromise between storage and fidelity, and it constrains what can be seen. Ten-minute averaging preserves the relationships that drive performance analysis and destroys the high-frequency content that classical condition monitoring relies on, which is why the two disciplines developed around different data rates. The standard deviation channel is the partial exception, because it retains a measure of within-interval variability and is the basis for computing turbulence intensity.

This is the raw material for performance work, and its great advantage is that it already exists. No new sensors, no site visit, and no interruption to operations are needed to begin - which is why an assessment can run entirely on history before any decision about deployment.

The practical difficulty is that SCADA arrives inconsistent. Manufacturers name channels differently, timestamps disagree between systems, gaps appear during outages, and status codes follow no shared standard. Getting a mixed fleet into one honest shape is most of the engineering, and everything downstream depends on it: a power curve learned from misaligned data describes the misalignment rather than the turbine, and predictive maintenance built on it inherits the same error.