Maiden Application of Skill Optimization Algorithm on Cascaded Multi-Level Neuro-Fuzzy based Power system Stabilizers for Damping Oscillations

Abstract

Power system oscillation is a significant threat among interconnected power systems that may lead to instability. The success of oscillation damping is primarily responsible for a modern power system's safe operation. However, the development of damping controllers is a multimodal optimization problem with constraints which challenges the traditional optimization algorithms. This paper critically examines damping schemes and controller stability analyses to find solutions to these issues and improve the performance of a multi-machine power system (MMPS) . This paper reveals the technique of improving power system stability by employing the Skill Optimization Algorithm (SOA) to optimize the gains of the conventional power system stabilizer (PSS) and the Neuro-Fuzzy inputs-output scaling parameters. This study aims to propose a multi-level Neuro-Fuzzy power system stabilizers (MLNFPSS ) to reduce the instability of a multi-machine two-area power system (MMTAPS) under fault conditions. The simulations were run on a MMTAPS considering a transmission line fault in the middle. The analysis leads to the deduction that the proposed MLNFPSS response is more efficient than conventional PSS under symmetrical three-phase faults. All control strategies have been executed, and the simulation results have been assessed using MATLAB 2016b/Simulink.