摘要：AbstractThe main contribution of this paper consists of the development of two methods for actuator fault estimation in dealing with the partially decoupled disturbances of the descriptor system, which is divided into decoupled and non-decoupled unknown inputs (UI). Based on the conventional UI observer, both of the solutions decouples the fault estimation with the first group of UI, while the second UI group is handled differently by each method. Finally, a numerical example with comparisons points out the performance of each approach.
