A New Frequency Control Method for Isolated Networks with Neural Network Controller Trained by PSO algorithm

This paper proposes a multilayer perceptron neural network (MLPNN) trained by particle swarm optimization (PSO) algorithm for frequency control of a new hybrid wind turbine (WT), photovoltaic (PV), fuel cell (FC) and ultra-capacitor (UC) for stand-alone applications. WT and PV are the primary power sources of the system, and an FC is expected to provide long-term energy balance, whereas the UC is employed as buffer storage for the short-term compensation. Changes in wind, solar light, and load consumption in isolated networks, prevent droop controllers and improved droop controllers to provide a proper performance over a wide range of operating conditions. To overcome this challenge, this paper proposes a new intelligent method by using a combination of MLPNN controller and particle swarm optimization (PSO) techniques for frequency controllers in isolated networks. To demonstrate the effectiveness of the proposed MLPNN controller, comparison with droop and improved droop controllers are performed using MATLAB/Simulink by integrating the detailed mathematical and electrical models of the hybrid isolated network

A New Frequency Control Method for Isolated Networks with Neural Network Controller Trained by PSO algorithm Keywords:

Frequency control , dynamic modeling , Isolated Network , Multilayer Perceptron Neural Network (MLPNN) , Improved droop control , Particle Swarm ptimization (PSO)