A Perception for Wide Area Power Monitoring System

This paper focuses on the advanced optimization technique which is used to know the dynamic behavior of the power system when it is subjected to any disturbance by tuning the data of the power system. Power system is a major complex system and therefore developing detail mathematical modeling is very difficult. That's the reason neural network is used to estimate the dynamic behavior of the power system, not from the mathematical modeling, but from the input-output data of the power system. A recurrent neural network (RNN) has been proved to be very effective for dynamic system identification. It is important to tune the weights of RNN, so that optimal values of those weights are obtained and the neural network can effectively predict the system states. In this project, particle swarm optimization technique-Quantum Infusion is used to reduce average error value so that it can track the system dynamics of the power system properly. Particle Swarm Optimization (PSO) is generally used to tune the weights of a neural network. PSO, though very effective, sometimes get trapped into a local optimum and fail to reach the global optimum. An advanced version of PSO which is called PSO with Quantum Infusion (PSO-QI) can be very effective to avoid being trapped in a local optimum. Finally, the dynamic behavior of the power system is demonstrated using the MATLAB software package.