Title :
Dynamic performance model of wind turbine generators
Author :
Gautam, Prajwal K. ; Venayagamoorthy, Ganesh K.
Author_Institution :
Holcombe Dept. of Electr. & Comput. Eng., Clemson Univ., Clemson, SC, USA
Abstract :
To characterize the power performance of wind turbine generators (WTGs), the International Electrotechnical Commission provides a power curve for each one based on ten minutes average. This approach causes the problem of systematic errors because of the nonlinear relationship between power and wind speed. In this paper, recurrent neural networks are introduced as an alternative approach to model this nonlinear relationship. Based on actual wind speed input and wind power output, an input-output relationship is established for a permanent magnet synchronous machine based wind power generator. Experimental studies are carried out to the develop power curve that characterize dynamic power performance of the wind turbine generator. These dynamic power performance models of wind turbine generators can be used as operational and planning models in control centers. Some preliminary results on the integration of a neural network wind generator model in a micro-grid simulation is presented.
Keywords :
distributed power generation; permanent magnet machines; power engineering computing; power generation control; recurrent neural nets; synchronous machines; turbogenerators; wind turbines; WTG; control centers; dynamic performance model; international electrotechnical commission; microgrid; permanent magnet synchronous machine; recurrent neural networks; wind power generator; wind turbine generators; Generators; Neural networks; Testing; Wind power generation; Wind speed; Wind turbines; dynamic neural network; dynamic performance model; micro-grid; particle swarm optimization; power estimation; wind turbine;
Conference_Titel :
Computational Intelligence Applications In Smart Grid (CIASG), 2013 IEEE Symposium on
Conference_Location :
Singapore
DOI :
10.1109/CIASG.2013.6611505