A novel faulted section location technique for future active distribution networks

Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2023
Thesis identifier
  • T16671
Person Identifier (Local)
  • 201871895
Qualification Level
Qualification Name
Department, School or Faculty
  • Distribution Network Operators (DNOs) face increasingly higher challenges to preserve quality and continuity of supply due to the widespread penetration of Distributed Energy Resources (DER) [1–8]. In parallel, more advanced technologies are being introduced into secondary substations for better observability and controllability. These features provided via instrumented substation assets and Information Communication Technologies (ICT) present opportunities for the development and implementation of new functions aiming to the effective operation and monitoring of active distribution networks [9–14]. This thesis focuses on one of these functionalities – that is, leveraging the ability of Low Voltage (LV) sensors to locate 11 kV unsymmetrical faults by monitoring and processing the network voltage profile during fault conditions. In particular, a novel technique has been developed which identifies the Faulted Section (FS) of the Medium Voltage (MV) feeder after a fault has occurred. The proposed algorithm, of which the successful operation depends solely on distributed LV voltage monitoring devices, represents the main contribution of the research work. A key characteristic is that, although the LV sensors connected at the secondary side of MV/LV step-down transformers require communication to transmit the data to a central point, they do not require time synchronisation. The technique facilitates the fault location procedure, which is of major importance as it accelerates restoration, reduces the system downtime, minimises repair cost, and hence, increases the overall availability and reliability of the distribution network. Moreover, the thesis deals with the challenges related to the complexity of modern distribution networks, taking into account ring topologies, MV lateral connections, pre-fault load unbalance and the presence of DERs. In this sense, the empirical characterisation of grid connection stability and fault response of small scale commercially available LV PV inverters was realised. The purpose was twofold: 1) highlight the diversity among the inverters’ responses as observed during the testing and indicate the risk of loss of PV generation during typical MV and HV level faults and 2) develop a dynamic model representing the behaviour of a real inverter under the applied physical testing conditions. The particular model was deployed in the power system studies conducted, aiding the evaluation of the FS location technique. Laboratory investigation was also carried out at the facilities of the Power Networks Demonstration Centre (PNDC) to further examine the performance of the developed faulted section location algorithm. The tests were performed in both MV radial and ring PNDC network configurations and measurements were acquired from various LV test-bays. It was demonstrated that the scheme can reliably identify the faulted section of the line while consistently maintaining high accuracy across a wide range of fault scenarios. Further sensitivity analysis demonstrates that the proposed scheme is robust against partial loss of communications and noise interference. The thesis concludes with an overview of future work that is required to further advance the concepts demonstrated.
Advisor / supervisor
  • Coffele, Federico
  • Campbell, Booth
Resource Type
Date Created
  • 2022