Improve Data Quality
Every data set has issues.
Quickly detect and flag problems such as tower shading, icing events, sensor malfunctions, or low signal-to-noise ratio. Filter bad data from calculations, apply scaling factors, offsets, or time shifts, and even fill gaps.
Flag data segments based on visual inspection
By simply clicking and dragging on a time series graph or a scatter plot, you can manually apply flags to data segments to identify problems such as sensor malfunctions or icing events. Once you have applied flags, you can use those flags to filter data from graphs and calculations.
Flag data segments based on formal criteria
Create your own flag rules to flag automatically all data segments that meet the criteria you specify. Flag suspected icing events, for example, by searching for instances of low wind speed standard deviation and temperature below freezing.
Automatically flag data affected by tower shading
Windographer automatically detects tower shading patterns and generates the corresponding flag rules, so with two clicks you can flag every data point affected by tower shading.
Combine colocated anemometers
The Combine Anemometers window automatically combines each pair of colocated anemometers into a single data stream that factors out the effects of tower shading. It generates one column for the combined mean wind speeds, and another for the combined standard deviations, and automatically specifies the properties and relationship of the new columns.
Extrapolate data vertically
Windographer can synthesize speed, direction, and temperature data for any height above ground by applying the shear, veer, and temperature gradient patterns in the current data set or in any other data set. For example, you can extrapolate a met tower data set using shear patterns calculated from a nearby LiDAR data set.
Windographer can fill gaps both by extrapolating valid measurements made at another height, or by generating an entirely artificial data segment in a time interval containing no valid measurements. The synthesized data segments replicate the important statistical characteristics of the observed data, including distribution, diurnal and seasonal patterns, and autocorrelation, and they preserve the observed patterns of shear, veer, temperature gradient, and turbulence.
Define your own calculated columns
You can define a wide variety of calculated columns including the average of, ratio of, or difference between other data columns; shear, veer, and temperature gradient parameters; polynomial and piecewise linear functions; moving averages; rotor-equivalent wind speed; and many solar variables such as solar azimuth and clear-sky radiation.
Scale, offset, time shift, delete
Convert units, correct a scaling error, or apply a directional offset to one or more data columns. Change time zones or correct clock errors by shifting any piece of the data set in time. Delete a particular data segment, or delete all data flagged with a particular flag. You can even add randomness to quantized data to improve the performance of MCP and extreme wind analysis processes.