Fault-Tolerant Control for SSSC Using Neural Networks and PSO

This paper presents a fault-tolerant indirect adaptive neuro-controller (FTNC) for controlling a static synchronous series compensator (SSSC), which is connected to a power network. The FTNC consists of a sensor evaluation and restoration scheme (SERS), a radial basis function neuro-identifier (RBFNI) and a radial basis function neuro-controller (RBFNC). The SERS is designed using the auto-associative neural networks (auto-encoder) and the particle swarm optimizer (PSO). This FTNC is able to provide efficient control to the SSSC when single or multiple crucial sensor measurements are unavailable. The validity of the proposed FTNC model is examined by simulations in PSCAD/EMTDC environment