ASSESSMENT OF CONSERVATION VOLTAGE REDUCTION ON HV AND LV DISTRIBUTION NETWORKS

Abstract

Conservation voltage reduction (CVR) is a technique which provides energy reduction by lowering the supply voltage without violating voltage limits. It has been traditionally applied on high voltage (HV) distribution networks to reduce the energy consumption and energy loss. Only in recent years, CVR has been studied on low voltage (LV) distribution networks. However, the implementation of CVR on LV networks increases the voltage and energy losses on the upstream HV networks. The increase of loads and generations of low carbon technologies (LCTs), such as electric vehicles (EV), has also interfered with CVR performance. This thesis proposes a Monte Carlo (MC) based CVR model, which is capable of coordinating and optimizing the control devices, minimizing energy and losses on both HV and LV networks and assessing the impact of different EV models on CVR performance. The coordinated CVR model is proposed, which coordinates the LV tap changer and capacitors over each pre-determined control interval and provides energy reduction and loss reduction in LV networks. The coordinated CVR model is implemented on 38 LV trial networks and an average of 5.2% energy reduction is achieved. The coordinated CVR model with three different EV charging models (i.e. uncoordinated charging, V2G charging and mixed charging) are implemented on typical LV networks. The results show that the impact of V2G charging on CVR energy reduction capability is the least, while the mixed charging can help CVR to increase EV penetration capability more effectively. The particle swarm optimization (PSO) based CVR model (PSO-CVR) is proposed, which optimizes the HV and LV tap changers and capacitors and minimizes the energy consumption and energy loss on both HV and LV networks. The results show that PSO-CVR reduces the total energy by 7.5%, which provides 45MWh daily energy saving for the entire HV/LV trail area. The PSO-CVR is also compared with the genetic algorithm (GA) optimisation. The results show that the PSC-CVR has a better energy reduction capability and a significant better computational efficiency.

Date of Award	1 Aug 2019
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Haiyu Li (Supervisor) & Luis(Nando) Ochoa (Supervisor)