INTEGRAL SLIDING MODE CONTROL BASED ON EXTREME LEARNING MACHINE FOR A WIND TURBINE

Miloud Koumir, Ayoub E. Bakri, and Ismail Boumhidi

Keywords

Sensor-less variable speed wind turbine, extreme learning machine,integral sliding mode control

Abstract

This paper presents an integral sliding mode controller (ISMC) based on the extreme learning machine (ELM) algorithm used for robust and intelligent control of a sensor-less variable speed wind turbine. The two control objectives, defined in the region where the wind speed is below its nominal value, are to maximize the energy converted and to maintain the security of the system. Thus, the proposed approach explores the effectiveness of the ELM for single hidden layer feed-forward neural networks (SHLFNN) in order to improve the model needed to build the control, and more particularly to estimate the uncertain nonlinear function modelling the nonlinear aerodynamic torque. The obtained hybrid model is then used by the ISMC for law control system synthesis. The purpose of the proposed technique is to overcome the problems of the reaching phase and the chattering caused by using traditional sliding mode control (SMC) approach, mainly for large uncertain systems due to the higher needed switching gain. The stability of the proposed (ISMC_ELM) control is analysed using Lyapunov criterion, and its efficiency is illustrated via simulations based on comparisons with the traditional SMC.

Important Links:



Go Back