This paper discusses the implementation of scientific techniques required to process Lidar data for the Wind Industry. Lidar is a remote sensing method that enables analysts to assess characteristics of the wind and infer critical information relating to the measured volume.
Lidar devices for wind power applications are typically Doppler Lidars which emit a laser beam and detect the Doppler shift of the back-scattered signal from aerosol particles carried by the wind. The measurements represent the radial velocities of particles along the line-of-sight. Spatial and vector information about the wind can be extracted using techniques such as Doppler Beam Swinging or Velocity Azimuth Display.
In this project we used Python and its scientific libraries to manipulate and apply the processing techniques to the measurements. The challenges addressed included processing of large amounts of data and designing a flexible and reusable framework to implement “dynamic models” of processing depending on the desired output. The building blocks we used were Python, Numpy, Scipy, Pandas, PyTables, Matplotlib, PureMVC and HTML5.