A novel adaptive power system stabilizer design using the self‐recurrent wavelet neural networks via adaptive learning rates. (24th February 2012)

• Record Type: Journal Article
• Authors: Ganjefar, Soheil; Alizadeh, Mojtaba
• Is Part Of: International transactions on electrical energy systems. Volume 23:Number 5(2013:Jul.)
• Page Start: 601
• Page End: 619
• Publication Date: 2012-02-24
• Subjects: power system stabilizer -- self‐recurrent wavelet neural network -- adaptive learning rates -- intelligent control; Electric power -- Periodicals; Electric power systems -- Periodicals; Electrical engineering -- Periodicals; 621.3
• DOI: 10.1002/etep.1616
• Languages: English
• ISSNs: 2050-7038

Abstract

SUMMARY: In this article, the self‐recurrent wavelet neural network (SRWNN) is used as a controller in both direct and indirect adaptive control structures to damp the low‐frequency power system oscillations when only the inputs and outputs of synchronous generator are accessible for measurement. The gradient descent method using adaptive learning rates (ALRs) is applied to train all weights of SRWNN. The ALRs are derived from the discrete Lyapunov stability theorem, which was applied to guarantee the convergence of the proposed control schemes. Finally, the proposed control schemes are evaluated on a single machine infinite bus power system under different operating conditions and disturbances to demonstrate their effectiveness and robustness. Copyright © 2012 John Wiley & Sons, Ltd.