A New Self-Tuning Deep Neuro-Sliding Mode Control for Multi-Machine Power System Stabilizer

In order to increase the accuracy and improve performance of power system stabilizer (PSS) controller compared methods presented in other studies, this paper presents a new method for tuning sliding mode control (SMC) parameters PSS using deep neural network. This requires fast switching which can create unwanted signals. To solve problem, boundary layer is used. First, equations multi-machine are converted into standard form control, then surfaces determined with three unknown parameters. Calculating determining optimal values (at any moment) these fundamental challenges. A network overcome challenge adjust regularly. simulation, 4 machines 11 buses implemented both phase-to-ground three-phase errors applied. The simulation results clearly show good proposed especially importance SMC structure methods.