Skip to Main Content
Recently, there has been a surge of interest in the development and use of flexible AC transmission systems (FACTS) controllers in power transmission systems. The most popular type of FACTS devices in terms of application is the static var compensator (SVC). In FACTS, there are many uncertainties such as sudden load shedding, generation tripping, occurrence of faults, change of parameters, and network configuration. Traditional linear controllers have many difficulties in treating these uncertainties. To overcome this problem, sliding mode control (SMC) has been widely used as one of the precise and robust algorithms. Although classical SMC is a powerful scheme for nonlinear systems with uncertainty, this control scheme has important drawbacks limiting its practical applicability, such as chattering and excessive control action. To alleviate the problems, the discontinuous part of the control signal in the classical SMC is substituted by neural network to eliminate the chattering phenomenon. The neuro sliding mode proposed in this paper consist of a one layered neural network whose activation functions are linear. The main working principle of the controller is minimizing a cost function which is determined from the requirements of the Lyapunov stability criteria and sliding mode control theory. Simulation results show the significant improvement of transient response which practically enables the use of a neuro sliding mode controller in FACTS.