Recurrent neural network based adaptive integral sliding mode power maximization control for wind power systems

• Published in: Renewable energy Vol. 145; pp. 1149 - 1157
• Main Authors: Yin, Xiuxing, Jiang, Zhansi, Pan, Li
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2020
• Subjects: Maximum wind power extraction; Recurrent neural network; Sliding mode control; Wind power system; Wind power system; Maximum wind power extraction; Recurrent neural network; Sliding mode control
• ISSN: 0960-1481; 1879-0682
• DOI: 10.1016/j.renene.2018.12.098

An adaptive integral sliding mode controller is proposed to maximize wind power extraction by maintaining the optimum rotation speed of wind turbine. In the proposed controller, an integral sliding mode control law is designed to track the optimum turbine rotation speed based on a recurrent neural network (RNN) that is used to identify the uncertain wind turbine dynamics. An online update algorithm is then derived to update the weights of the RNN in real time and hence to facilitate the maximum power extraction control. The stability of the overall control system is guaranteed in the sense of Lyapunov stability theory. Comparative experimental results demonstrate that the proposed controller outperforms a conventional control method in tracking the optimum turbine rotation speed and extracting the maximum wind power despite system uncertainties and high nonlinearities. •An integral sliding mode control law is designed to extract maximum wind power.•Recurrent neural network is used to identify the uncertain wind turbine dynamics.•An online update algorithm is derived to update the weights of the RNN.