Islanding Detection in Distributed Generation By Neuro-fuzzy Approach

The islanding methods are classified in terms of the islanding principle and the distributed Generation (DG). Islanding detection techniques are mainly divided into three types for distribution systems: Remote, local and communication based techniques. Passive methods are based on the information available on the DG site at the point of common coupling (PCC) with the utility grid. A new approach in passive techniques is the use of data-mining to classify the system parameters. Passive techniques are fast and they don’t introduce disturbance in the system but they have a large non detectable zone (NDZ) where they fail to detect the islanding condition. In this paper, considerable indices of a distribution system are collected by using MATLAB simulation. These indices are change in power, change in voltage, rate of change of power, rate of change of voltage, total harmonic distortion (THD) current, voltage, and change in power factor. Adaptive Neuro-Fuzzy inference System (ANFIS) in MATLAB used to classifying for these indices and define the boundaries. The results show the use of ANFIS in reducing the NDZ of passive islanding detection systems. KeywordsNon detectable zone (NDZ), Adaptive NeuroFuzzy inference System (ANFIS), Distributed generation (DG), islanding detection.