Echo-state-network-based real-time wind speed estimation for wind power generation

Authors:
Wei Qiao
Affiliations:
Department of Electrical Engineering, University of Nebraska-Lincoln, Lincoln, NE
Venue:
IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Year:
2009

Citing 5
Cited 0

Multilayer feedforward networks are universal approximators

Neural Networks
Layered neural networks with Gaussian hidden units as universal approximations

Neural Computation
Universal approximation using radial-basis-function networks

Neural Computation
2007 Special Issue: Optimization and applications of echo state networks with leaky- integrator neurons

Neural Networks
A learning algorithm for continually running fully recurrent neural networks

Neural Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Wind turbine generators (WTGs) are usually equipped with one or more well-calibrated anemometers to measure wind speed for system monitoring, control, and protection. The use of these mechanical sensors increases the cost and hardware complexity and reduces the reliability of the WTG system. This paper proposes an echo-state-network (ESN)-based real-time wind speed estimation algorithm for WTG systems. The ESN is designed to provide a nonlinear inverse model of the WTG dynamics, which is used to estimate the wind speed in real time from the measured WTG output electrical power and shaft speed at any turbine blade pitch angle. The estimated wind speed is then used for wind-speed-sensorless control of the WTG system. The proposed algorithm is verified by sim ulation studies on a 3.6-MW wind turbine equipped with a doubly fed induction generator (DFIG).