Voltage Collapse Prediction By Neuro-Fuzzy (ANFIS) Scheme

In recent years, number of voltage stability indices have been suggested for the voltage collapse appraisal.Many of them are extracted by very complex analytical tools and are unmanageable to be rendered by the system operators. In this work, an Artificial Intelligence (AI) based approach has been used that integrate the strengths of Fuzzy Logic (FL) and the Artificial Neural Network (ANN). A decision model built on FL takes as an input a given set of voltage stability indices (numerical values), representing a snapshot of the actual operating point for the electric system.The set of numerical values is read into a set of symbolic and linguistic criteria. These variables are wangled by a set of logical connectives and of Inference Methods offered by theMathematical Logic. By this way, the FL definesa metric in terms of a percentage rate of the security level degradation with respect to the voltage collapse risk [1]. In this investigation ANFIS model is used to predict the stability margin or distance to voltage collapse and hence security status of power system network based on reactive power load demand by feeding four predefined voltage collapse indices as input to the neuro-fuzzy system.The proposed technique is applied to IEEE 14 and IEEE 30-bus test systems and got encouraging results.