Fault Reconstruction of Islanded Nonlinear DC Microgrids: An LPV-Based Sliding Mode Observer Approach

This article investigates the problem of fault detection and reconstruction in direct current microgrids (dc MGs) with nonlinear loads. First, a state-space representation of the dc MG with nonlinear constant power loads (CPLs) is obtained. The faults in components, actuators, and sensors in a dc MG system are modeled by using additive terms in the state-space and measurement equations. A novel robust nonlinear detector is then proposed to estimate and reconstruct the faults. The proposed detector deploys a sliding mode technique and a polytopic linear parameter varying (LPV) approach. The proposed approach utilizes the robustness and simplicity offered by the sliding mode and polytopic LPV approaches to provide a simple, but effective, detector. By employing the Lyapunov stability theory, sufficient detector design conditions are derived in terms of linear matrix inequalities (LMIs) which are numerically solved by convex optimization techniques. The proposed approach is experimentally tested on a practical dc MG benchmark connected to a CPL. Furthermore, the results are compared with other state-of-the-art methods.