ADAPTIVE ARTIFICIAL NEURAL NETWORK-BASED CONTROL OF WIND TURBINES

Maksim Karabanov, Arūnas Andziulis, Žydrūnas Lukošius, Denisas Bykovas

Abstract


In order to be economically competitive, various control systems are used in large scale wind turbines. In this paper, we propose anadaptive control based on radial-basis-function neural network (NN) for different operation modes of variable-speed variable-pitch windturbines and smooth transition between these two modes. The robust NN weight updating rules are obtained using Lyapunov stability. Theproposed control algorithm is first tested with a simplified mathematical model. Simulation results for single and multi-step predictionswere obtained using the MATLAB Simulink toolbox.

KEY WORDS: wind turbines control, neural networks, Matlab.


Full Text: PDF

Refbacks

  • There are currently no refbacks.