Improving Efficiency of load frequency control for smart grid using adaptive neuro-fuzzy inference system

Abstract

This research presents the Improving Efficiency of a load frequency control for smart grid ANFIS. In this research, the smart grid that uses electrical energy from Steam turbine power plant PV arrays and wind turbines generator was controlled by ANFIS. Processes of ANFIS include 1) predicting the total power in the system by neuron network and 2) select the appropriate control gain for the power system in various situations by fuzzy logic. For validating the performance of the proposed method, the results from the proposed method are compared with the conventional method that is used to control the load frequency using the PI controller and load frequency control using PID controller. The simulation is divided into 2 cases: 1) the power is continuously changing, 2) the power is changed momentarily. Furthermore, all the simulations are performed in MATLAB/Simulink. From the simulation results, efficiencies of all cases have the same as the trend that the ANFIS controller is fastest effective in making the system back to the stability and has the lowest amplitude of the changed frequency. The proposed method decreases amplitude 86% and decreases settling time 27.5% from the PI controller. Furthermore, the proposed method decreases the amplitude 35.4% and decreases settling time 15.8% from the PID controller.