Allocation of fault indicators in distribution feeders containing distributed generation

Abstract

As distribution systems are typically radial and branched, different branches have the same accumulated impedance from substation. Consequently, the impedance-based distance estimation techniques may identify multiple suspected locations for the same fault. The allocation of fault indicators reduces this problem. However, with distributed generation in distribution systems, the fault current, previously fed only by the substation, is now also fed by distributed generators. This may cause an incorrect operation of conventional fault indicators, requiring directional ones. In this context, an approach for allocation of conventional and directional fault indicators in distribution feeders, taking into account the distributed generation, is proposed in this paper. To represent the distance traveled by the maintenance teams during faults location, the proposed approach uses actual paths between the suspected fault locations, making the method realistic. Furthermore, using a NSGA-II algorithm, the best set of conventional and directional fault indicators required is determined. Results show that conventional fault indicators work accurately in the presence of low power distributed generators (less than 20% of the substation power) and, in the presence of high power generators, few directional fault indicators are needed.