OPTIMAL DESIGN AND PLANNING OF AN INTEGILGENT DISTRIBUTION NETWORK

Abstract

Climate change is currently a public concern. As a significant amount of electricity is still produced from fossil fuels, there is a lot of work focusing on the decarbonisation of the electric industry. By 2020, it is expected that 15% of the total electricity demand in the UK is required to be generated from renewable energy sources (RES) with a 20% reduction in greenhouse gas (GHG) emissions. A series of efforts have been made in distribution systems for achieving the above targets such as widespread utilisation of distributed generations (DGs) and the application of electricity tariff schemes. However, a high penetration of the DGs results in the potential mismatch between the generation and the demand profile since the generation of typical DGs such as wind turbines (WTs) and photovoltaic (PV) largely depend on the availability of natural resources. The work reported in this thesis focused on the design of innovative network planning frameworks and investment models assuming the dynamic DUoS and variable RES generation pricing mechanisms. In particular, an optimum day-ahead pricing mechanism was developed for DGs and DNOs to manage their RES to achieve revenue reconciliation and connected more DGs. Based on the proposed DUoS charge approach and the principle of electricity price elasticity, an innovative dynamic-pricing framework was introduced to find out the optimum price signal with the consideration of wholesale market, benefits of power suppliers and customers’ behaviours. Moreover, a hybrid optimization approach was introduced to solve the optimal conductor size selection (CSS) problem in the distribution systems planning process with the high penetration of DGs. Finally, a genetic algorithm based distribution network reconfiguration approach was designed to provide the optimum day-ahead network topologies considering the predicted data of load demand and available output of the distributed generations. The numerical results demonstrated that the proposed day-ahead pricing mechanism efficiently reduce the mismatch between the RES generation and demand profiles and relieve the network congestion problems during both peak demand or peak generation periods. Moreover, DNOs can use the proposed framework to provide the optimized time-based electricity price to their customers for changing the time and amount that customers consume the electricity, thus avoiding the curtailment of RES or high DUoS charge. The numerical results also demonstrated that the proposed CSS approach can allocate the suitable conductor type from the given inventories to each branch in the network.

Date of Award	31 Dec 2019
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Joseph Mutale (Supervisor) & Pierluigi Mancarella (Supervisor)