A feedback linearization based fuzzy-neuro controller for current source inverter-based STATCOM

This paper presents a nonlinear control approach to the MIMO system of a current source inverter (CSI) based static synchronous compensator (STATCOM). Nonlinear control approach based on feedback linearizing scheme for FACTS devices has been proved in previous literature to have superior performance In damping the electromechanical oscillations of the power system. In this proposed control scheme, the artificial neural network (ANN) will be trained based on feedback linearization control scheme. Radial basis function neural networks (RBFNN) are used as online approximators to learn the unknown dynamics of the system. However, steady state error after the disturbances occurs in conventional feedback linearizing controller. Thus, training of RBFNN as conventional feedback linearizing controller became unrealizable. Consequently, fuzzy controller based on TSK IV control scheme has been used to filtered out the steady state error. This proposed controller is expected to approximate and replace complex mathematical equations of feedback linearization control scheme. To demonstrate the application of the proposed controller, case studies are done with a single-machine infinite-bus power system with current source inverter (CSI)-based STATCOM installed at certain bus. The efficacy of the control strategy is evaluated by digital computer simulation studies using MATLAB under various transient disturbances and a wide range of operating conditions. The approximated control signals are compared with that of targeted control signals to exhibit the elegance of the proposed control scheme.