RBF neural networks sliding mode controller design for static var compensator

To enhance the transient stability of the electric power control system, a radial basis function (RBF) neural networks sliding mode controller design is proposed for static var compensator (SVC) with uncertain parameter. Unlike the conventional adaptive control schemes, the certainty equivalence principle is not required for estimating the uncertain parameter in adaptive law design. Based on the system immersion and manifold invariant (I&I) adaptive control, the designed adaptive law ensure that the estimation error can converge to zero in finite time. In addition, the control law is designed by the (radial basis function) RBF sliding mode control. The neural networks can compensate for the nonlinear uncertain effect in SVC system by its universal approximation ability. The effectiveness of the proposed controller is verified by the simulations. Compared with adaptive backstepping sliding mode and adaptive backstepping, the oscillation amplitudes of system state variables are reduced by at least 17%, and the response approaches steady state is shortened by 7%.