Wind prognosis for wind power devices

(for better viewing click the pictures please)

  • Prognosis system for determination of the wind velocity and the energy input of wind power devices within the short time range
  • Employment of newest mathematical procedures: Wavelet analysis, Kalman filter
  • Wind prognosis by itself sequentially optimizing neural nets => adaptive prognosis system
  • variable parameter settings of the prognosis models
  • Library of neural nets
  • Multithread application
     Order number
  • 30435

The wind prognosis module serves the prognosis of the wind velocity and the energy input of wind power devices within the short time range. The module makes further functions available to the representation and evaluation of the prognosis results.

Data base
For the wind prognosis meteorological data is needed seized with the software UK_TOSS seized meteorological data is needed in 15 minute raster. Therefore the wind prognosis module can be used only in connection with UK_TOSS. The possibility of the linkage of the meteorological data of different weather stations by means of a data supply matrix permits the consideration of local connections in a network from weather stations.

Neural networks
For the prognosis of the wind velocity primarily neural nets are used. Other mathematical procedures serve this for the support. For everyone time interval which can be prognosticated an own prognosis process is created.

Wavelet analysis
Time series are divided by the Wavelet analysis in the time frequency area. In the wind prognosis module the results of the Wavelet analysis get as on-line parameters into the neural net, in order to improve its forecast quality. The cross correlation matrix of the analyzed meteorological data can be plotted.

The diagram offers an overview of the wind prognoses of all weather stations. It combines a graph of the measured and prognosticated wind velocities with a tabular representation of these values as well as additional information like the middle error. In addition for each prognosis time interval an error curve is indicated.

Tabular overview
The tabular representation offers an overview of all defined prognoses of all weather stations. Measured ones and prognosticated values of the wind velocity are coloured differentiated.

Wind prognosis status
A status window gives a global overview of all up-to-date running prognosis processes. The details of a process can be displayed by its selection.

Energy characteristic
The connection between the prognosticated wind velocity and the energy input which can be expected of the wind power devices concerned is given by an energy characteristic. It consists of pairs of electric power produced at wind velocity in steps of 1 m/s and must be entered individually depending upon wind power device.

further explanations
please read our report on the development here (in German)


     System requirements
Software UK_TOSS executably installed on PC

Impressum | Privacy Policy | Copyright TOSS GmbH i.L. Potsdam
For questions and comments referring to this website email to: TOSS GmbH i.L.
Last update: 20.06.2024