Protection strategy in active DC power distribution networks

Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2020
Thesis identifier
  • T16105
Person Identifier (Local)
  • 201562267
Qualification Level
Qualification Name
Department, School or Faculty
  • Environmental incentives to combat climate change are providing the motivation to improve the energy efficiency of power distribution systems and integrate state-of-the-art renewable technologies. DC distribution networks are receiving considerable attention in the literature because they offer a simple and flexible interface between these modern resources and consumers. However, many technical challenges relating to the design and standardisation of DC protection devices still exist that must be overcome prior to widespread adoption. Since DC fault current develops rapidly, many high-speed protection schemes tailored for DC networks have been proposed. However, few of them have considered the difficulties in practical implementation. This thesis will present the implementation challenges and propose corresponding protection schemes to address the issues. In seeking to achieve this aim, the work presented within this thesis makes three main contributions. This thesis has firstly improved the reliability of the high-speed DC differential protection scheme. The main implementation challenge of this scheme is that a short time synchronisation error may cause a significant current difference error, resulting in a false-trip problem when a fault occurs outside the protected zone. This thesis has proposed a "multi-sample differential (MSD) protection scheme" to ensure the protection stability for external zone faults (i.e., the relays must not operate) whilst maintaining sensitivity for internal zone faults (i.e., the relays must operate) by examining multiples measurement samples. Secondly, the difficulty in realising high-speed DC distance protection is that measurement of rate-of-change of current can be severely affected by even low-level noise, resulting in a failure in fault detection. This thesis has presented the methodology for selecting the appropriate sampling time of the numerical derivative as well as the cut-off frequency of low-pass current measurement lfiters. Although high-speed protection schemes can effectively isolate faults quickly, their implementation requires many advanced devices, which may not be economical for lowpower and low-cost DC networks. Finally, this thesis has proposed a "modulated low fault-energy (MLE) protection scheme" that employs fault current limiters (FCL) at the grid energy sources and mechanical circuit breakers (MCB) elsewhere throughout the distributed network. This deployment can constrain the fault current to a lowenergy level that enables a longer time window for the downstream MCBs to realise protection with a lower total implementation cost. Drawing conclusions from this PhD research, the author advocates that more consideration should be given to implementation challenges when designing protection schemes in DC distribution networks. Excessive pursuit of ultrafast fault isolation speeds can lead to over-cost and protection instability issues in practice. A prospective protection scheme must compromise between the high-speed protection requirements in theory and the reliable but economical requirements in practice, to accelerate the realisation of large-scale DC grids in future.
Advisor / supervisor
  • Norman, Patrick
Resource Type