Neural network based recursive terminal sliding mode control and its application to active power filters

In recent times, research interests for the control design of uncertain nonlinear systems are observed in neuroscience-inspired intelligent controllers. In this article, a recursive terminal sliding mode control scheme based on continuous radial basis emotional neural network (CRBENN) is proposed. First, a recursive terminal sliding mode controller is constructed. The sliding mode surface is composed of a fast non-singular terminal sliding mode surface and a recursive integral terminal sliding mode surface, which not only ensures that the tracking error converges to zero in a finite time, but also can eliminate the reaching phase and achieve global robustness by setting the surface parameters with appropriate initial values. In addition, in order to effectively deal with the uncertainty, an adaptive CRBENN controller with online parameter adjustment capability is established, and its stability and convergence are analyzed using the Lyapunov method. The designed CRBENN, inspired by neuroscience, is simple in structure and fast in response. Simulation and experimental results on active power filter show that the control scheme has good current tracking ability and anti-interference ability.