Cloud Neural Algorithm based Load Frequency Control in Interconnected Power System

Abstract

In order to effectively increase the control quality of the load frequency controller, this paper proposes a Cloud-Neural Network PI controller based on cloud theory to solve the adverse effects of uncertainties in interconnected power system. Cloud theory solves uncertain problems by skillfully combining probability statistics with fuzzy theory, but relies heavily on artificial experience, while proposed cloud neural network algorithm can learn cloud control rules by itself. Therefore, cloud-neural network PI controller can solve the problem of fixing the membership functions of input variables and fuzzy rules by clouds algorithm, and implement the nonlinear mapping between variables by neural network. Compared with cloud theory, this new algorithm retains self-learning function of the neural network and does not depend on artificial experience. On the Matlab platform, comparison of Cloud-Neural Network PI controller, Cloud PI controller and traditional PI controller is made in complex nonlinear power system, and simulation results show that proposed Cloud-Neural Network PI controller has strong adaptive and self-learning capabilities, presenting better robust performance and dynamic-static characteristic.