Nonlinear model identification and control of wind turbine using wavenets

In this paper a PI control strategy using neural network adaptive RASP1 wavelet for WECS's control is proposed. It is based on single layer feed forward neural networks with hidden nodes of adaptive RASPl (rational functions with second-order) wavelet functions controller and an infinite impulse response (IIR) recurrent structure. The IIR is combined by cascading to the network to provide double local structure resulting in improving speed of learning. This particular neuro PI controller assumes a certain model structure to approximately identify the system dynamics of the unknown plant (WECSs) and generate the control signal. The results are applied to a typical turbine/generator pair, showing the feasibility of the proposed solution

[1]  Miguel Angel Mayosky,et al.  Direct adaptive control of wind energy conversion systems using Gaussian networks , 1999, IEEE Trans. Neural Networks.

[2]  N.K. Loh,et al.  Dynamic System Identification using Recurrent Radial Basis Function Network , 1993, 1993 American Control Conference.

[3]  M. Sedighizadeh,et al.  Adaptive PID control of wind energy conversion systems using wavenets , 2004, 39th International Universities Power Engineering Conference, 2004. UPEC 2004..

[4]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[5]  N. D. Hatziargyriou,et al.  A new control scheme for variable speed wind turbines using neural networks , 2002, 2002 IEEE Power Engineering Society Winter Meeting. Conference Proceedings (Cat. No.02CH37309).

[6]  Shuhui Li,et al.  Using neural networks to estimate wind turbine power generation , 2001 .

[7]  M. Sedighizadeh,et al.  Adaptive PID control of wind energy conversion systems using RASP1 mother wavelet basis function networks , 2004, 2004 IEEE Region 10 Conference TENCON 2004..

[8]  Gaviphat Lekutai,et al.  Adaptive Self-Tuning Neuro Wavelet Network Controllers , 1997 .