Adaptive nonlinear control of three-phase shunt active power filters with magnetic saturation

Abstract

Abstract The problem of controlling three-phase shunt active power filters (SAPF) is addressed in presence of nonlinear loads. Previous works generally design control for SAPF based on standard models that assume the involved magnetic coil to be linear. In reality, the magnetic characteristics of these components are nonlinear (especially in the presence of large magnetic flux density in the ferromagnetic core). In this paper, a new oriented control model for SAPF-load system, taking into account for the nonlinearity of coil characteristics, is developed. The control objective is twofold: (i) compensating for the current harmonics and the reactive power absorbed by the nonlinear load; (ii) regulating the inverter DC capacitor voltage. To this end, based on the new model, a nonlinear controller is developed, using the backstepping technical design. It is therefore able to ensure good performances over a wide range of variation of the load current. Moreover, the controller is made adaptive for compensating the uncertainty on the switching loss power. The performances of the proposed adaptive controller are formally analyzed using tools from the Lyapunov stability and the averaging theory. The supremacy of the proposed controller with respect to standard control solutions is illustrated through simulation.