Automatic Generation Control (AGC) of Wind Power System: An Least Squares-Support Vector Machine (LS-SVM) Radial Basis Function (RBF) Kernel Approach. Issue 14 (14th September 2018)

• Record Type: Journal Article
• Authors: Sharma, Gulshan; Nasiruddin, Ibraheem; Niazi, K. R.; Bansal, R. C.
• Is Part Of: Electric power components and systems. Volume 46:Issue 14/15(2018)
• Page Start: 1621
• Page End: 1633
• Publication Date: 2018-09-14
• Subjects: automatic generation control, DFIG support wind turbines, robust control, ANN, SVM, LS-SVM, RBF kernel -- ANN -- SVM -- LS-SVM -- RBF kernel; Electric machinery -- Periodicals; 621.3104205
• DOI: 10.1080/15325008.2018.1511003
• Languages: English
• ISSNs: 1532-5008

Abstract

Abstract: The present paper discusses the structural, operational and control complexity of present day modern power systems in the wake of addition of electrical energy from wind turbines. The automatic generation control (AGC) problem of interconnected power system incorporating doubly fed induction generator (DFIG)-based wind turbines has been formulated and the investigations under various system operating conditions are presented. A two-area power system with DFIG-based wind turbines in each area are considered for the investigations. The system non-linearities such as governor dead-band and generation rate constraint are incorporated in the system dynamic model development. A novel AGC scheme using a non-linear least squares-support vector machine is proposed in the work. The proposed regulator is trained using a reliable data set consisting of wide range of operating conditions and area load changes generated by robust control technique. The results obtained with proposed regulator are compared with that achieved with multi-layer perceptron (MLP) and conventional PI regulators under various system operating conditions. The investigations carried out in the work have effectively demonstrated the superiority over MLP and conventional PI-based AGC regulators.