Identification of exact fault location on a distribution network using combined fault location and heuristic-based sensor placement algorithm

Fault occurrence on power system lines is an inevitable phenomenon. Therefore, mitigation of the resulting consequences of power interruption from such events becomes a priority. The process involved in remedying the faulted element on the distribution network consists of first, identifying the occurrence of faults on the system and, secondly, but more importantly, identifying the location of the faulted element swiftly and accurately. The impedance-based fault location (IBFL) techniques have been traditionally employed to estimate the location of the fault on a distribution system. These methods use system parameters to determine the distance to the fault on the network. Although straightforward when applied to simple circuits, these methods have an acceptable margin of error associated with them for approximating the exact location of the fault on the system. Furthermore, the accuracy of IBFL methods in approximating the exact location of the fault significantly reduces when implemented on an intricate distribution system. The reason behind this complication can be attributed to the complexity of the distribution network that arises due to the multiple laterals on the system. The IBFL methods result in the distance estimate of the fault from one end of the system instead of determining the exact location of the fault on the network. Therefore, the interpretation of the results to truly identify the fault's location is not straightforward because of the multiple laterals. One way to compensate for this limitation is to place additional sensors on the distribution system that could help interpret the distance estimate to identify the exact location of the fault. Hence, to overcome the limitation, a combined algorithm is proposed in this thesis. The proposed combined algorithm integrates the traditional impedance-based fault location methods like the Takagi method with an intelligent sensor network across the distribution feeder determined by the proposed sensor placement algorithm. When implemented with the Takagi method, the heuristic-based optimal sensor placement algorithm can elevate the accuracy of determining the exact fault location on the distribution network. The heuristic-based sensor placement algorithm is developed to determine the optimal location to place additional sensors on the system to achieve more accurate fault location. The results from the sensor placement algorithm are then used along with the Takagi method to find the exact fault location on the distribution network with more accuracy. The proposed sensor placement algorithm is developed using a test 15-bus distribution system and later implemented on a real-world distribution system to verify its application in better identifying the exact fault location established on the results from the IBFL method.