Kiprakis, Aristides

Harrison, Gareth

Wei, Mingyue

other

2021-11-12T10:40:35Z

2021-11-12T10:40:35Z

2021-11-27

High penetration of distributed generation (DG) is evident in the development of modern power systems, and this trend is expected to continue growing. Grid integration control strategies have been recognized as the main way to save investment, reduce fossil fuel energy consumption, and improve the reliability and flexibility of power systems. The distribution network itself is also a weak link that causes the deterioration of power quality and affects the overall performance and efficiency of the system. Therefore, smart distribution networks play a decisive role in the overall concept of the smart grid. Smart inverter control is more appreciated in the field of optimal operation of low voltage distribution networks with high penetration of distributed generations, due to its advantages of high flexibility, quick response time and low additional system cost. The work presented in this PhD thesis aims to investigate the impact of high penetration of distributed energy resource in the power system and develop a multi-objective optimisation method to improve the distributed network performance. This research studied the impact of the access of a large number of distributed energy sources such as rooftop solar photovoltaics on low carbon power availability and quality of supply within the distribution network. By looking for the best model composition, it created a distribution network model with highly distributed photovoltaics. The optimized maximum capacity and location of the solar energy configuration in the distribution network are realized through voltage sensitivity and genetic algorithm. The concept of adjusting reactive power output in distribution network to affect system characteristics are based on the smart inverter functions of distributed photovoltaics. To further conduct the multi-objective control, this thesis proposed an adaptive particle swarm optimisation scheme. The main algorithm based on multi-objective particle swarm optimization (MOPSO) algorithm, which uses the concept of Pareto dominance to find solutions for multi-objective problems. Through the investigation of the interactions of multiple effects such as voltage fluctuations, system cost and CO2 emissions in a distributed network with saturated PV penetration, with the advanced algorithm in planning photovoltaics and further smart inverter volt/var control scheme, the generation cost and GHG emission curve were improved and avoid the disturbance to customer behaviours. This thesis also proposed an adaptive multi-objective control scheme for distribution networks with photovoltaics infeed, on the basis of smart inverter control. To cater for the fact that low voltage distribution network may face problems such as voltage mitigation, system loss increase and customer also requires to pay less bill. The method is designed to solve these multiple conflict objectives. The correctness and validity of the proposed strategy are verified by numerical examples in a radial large 149 bus system. This thesis conducted detailed investigation of highly distribution network characteristics and proposed a process of multi-objective control scheme for the distribution network. It not only illustrated the control method, but also included the implementation requirement and consideration beforehand and afterwards. In addition, unlike mostly previous work, the work presented in this thesis included benefits of users as one of the optimisation goals. Therefore, this work can be adopted and applied by DSO as a technical reference and for trade-off decision makings in a highly distributed energy network.

https://hdl.handle.net/1842/38260

http://dx.doi.org/10.7488/era/1526

en

The University of Edinburgh

optimization system

photovoltaic

voltage quality

harmonic interference

technical standards

PV access

PV pre-configuration access

distribution network

reactive power compensation function

Multi-objective optimisation of highly distributed power system management

Thesis or Dissertation

Doctoral

PhD Doctor of Philosophy