Abstract
Purpose The problem of stability is generally caused by insufficient damping of electromechanical oscillations (EMOs). Power system stabilizers (PSSs) are the most advised and efficient devices to increase the system damping and enhance the dynamic characteristics of power networks during abnormal conditions. Unfortunately, the performance of the PSS controller is mostly dependent on the parameters of the lead-lag compensator. Within this context, this study presents a new chaotic-based sunflower optimization algorithm with local search (CSFO-LS) for optimum design of PSS controllers. Methodology In the proposed algorithm symbolized by CSFO-LS, the random parameters of the original SFO are substituted by chaotic sequences to avoid premature convergence at local optima and improve the accuracy of the optimum solution. Firstly, the CSFO-LS is tested and evaluated on various benchmark functions with different characteristics such as multimodality, separability and regularity. Then, it is applied for selecting the optimum parameters of the PSS controllers. These parameters are tuned in order to shift all electromechanical modes in a pre-specified zone in the left side of thes-plan. Results Simulation results based on eight benchmark functions show that CSFO-LS outperforms all the algorithms used for comparison. Moreover, to demonstrate the applicability and performance of the proposed method for providing good damping of low frequency oscillations, a standard power system test under various operating conditions and severe fault is used. Obtained results are compared with those obtained using the original SFO and other recent optimization techniques. Originality In this study, an improved version of the SFO is proposed for providing optimum EMOs damping. All EMOs have to be shifted as much as possible to the left side of thes-plan instead of shifting them to a fixed zone. To our best knowledge, this technique is not suggested or used for any power system problem.