DocumentCode :
2591319
Title :
Online knowledge based wind assessment tool
Author :
Octavian, C. ; Ioan, L.
Author_Institution :
Renewable Dept., IPA Cluj Napoca Subsidiary, Cluj Napoca, Romania
fYear :
2012
fDate :
24-27 May 2012
Firstpage :
352
Lastpage :
355
Abstract :
The main problems of current site prospecting techniques as they occur in real scenarios is referring the disadvantages of the mesoscale mathematical model, based on which the wind assessment uncertainties are appreciated around 30%, just by the wind analyst [1]. Once we propose to change the running mathematical models for wind potential estimation with a hybrid method based on expert rules and objective facts provided by the interested users as direct observations. For the renewable energy domain this concept is relatively new in academic literature and not used in applicative research yet. The rules are extracted from experts and good practices reports, whilst the inference engine is fed with human observations regarding temperature, altitude, pressure, obstacles, distances, position, asymmetry degree of the vegetation, etc. Part of them will be pre-processed with metric relations for checking them against the rules premises. Secondly the knowledge based system will run online, being accessible by the interested users all over the world.
Keywords :
geophysical prospecting; inference mechanisms; power engineering computing; wind power; expert rules; hybrid method; inference engine; mesoscale mathematical model; online knowledge based wind assessment tool; renewable energy domain; site prospecting techniques; wind assessment uncertainties; wind potential estimation; Cognition; Humans; Knowledge based systems; Measurement; Wind speed; Wind turbines; AI; Knowledge; wind assessment;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Automation Quality and Testing Robotics (AQTR), 2012 IEEE International Conference on
Conference_Location :
Cluj-Napoca
Print_ISBN :
978-1-4673-0701-7
Type :
conf
DOI :
10.1109/AQTR.2012.6237731
Filename :
6237731
Link To Document :
بازگشت