TY - JOUR ID - TI - Identification and Localization of non-zero Resistance Short circuit Faults in Distribution Feeders Based on the Theory of Wavelets and Artificial Intelligence AU - Sara J. Authafa ساره جعفر عذافة AU - Khalid M. Abdul-Hassan خالد مهدي عبدالحسن PY - 2017 VL - 17 IS - 2 SP - 18 EP - 32 JO - Basrah Journal for Engineering Sciences مجلة البصرة للعلوم الهندسية SN - 18146120 23118385 AB - This paper introduces a radial distribution feederprotection scheme based on certain features extraction fromcurrent signals measurement at the substation. The featuresare captured using the discrete wavelet transform (DWT). Twodigital signals processing methods are used to introduce thosefeatures to the 1) fault detection 2) identification and 3)localization schemes; the first one is the energy method and thesecond one is the root mean square method. For the purpose offault type identification, two systems are tested and compared,a Fuzzy Inference System (FIS) and Artificial Neural Network(ANN). Fault location scheme is then built based on ANNs. Aneffort is made to reduce the computational burden and thespeed of detection provided by the fault detection andidentification schemes. Since the short circuit faults are themost likely types of faults that can occur in power systems, theten types of these faults taking into account different faultresistances are simulated in MATLAB environment and theprotection scheme is built based on the idea of overcurrent.The power quality disturbances such as switching transientsevents on the feeder are also taken into account in order to builda reliable and secure protection scheme.

ER -