Cost reflective network pricing for high voltage and low voltage distribution networks

Abstract

Within the context of privatisation and deregulation in power industry, network users such as demand and generation should pay for their use of the networks, which always takes the form of Use-of-System (UoS) charges. Distribution network pricing is crucial in playing two roles: 1) ensuring economic efficiency, i.e. sending price signals to inform users of the network with respect to the costs they impose on networks and to influence the future behaviours of prospective users for efficient utilisation of existing networks; 2) enabling Distribution Network Operators (DNOs) to recover network investment costs.

In the UK, Distribution Reinforcement Model (DRM) has been the foundation of the network charges setting since the early 1980s. However, this approach neither reflects the extent of use of the network by users, nor provides price signals to influence the behaviours of users in the network. Hence, lack of economic efficiency of the DRM makes it necessary to develop new charging mechanisms for distribution networks.

The reform has been undergoing for Extra High Voltage (EHV Distribution-132kV, 33kV and 22kV in the UK) distribution networks, for which two new charging models are considered as the best available approaches by industry to achieve high level of economic efficiency, i.e. Long Run Incremental Cost (LRIC) and Forward Cost Pricing (FCP). However, the current DRM pricing model is retained for the High Voltage (HV Distribution -11kV and 6.6kV in the UK) and Low Voltage (LV Distribution-0.4kV in the UK) network charging because of: 1) the complexity of the two models; 2) the extensiveness of network configuration and limited available data.

The main objective of this thesis is to propose new charging models to achieve economic efficiency for HV and LV distribution networks. On the one hand, the key cost drivers need to be assessed properly as charges should be levied in line with what drives network costs. In this study, thermal and voltage constraints are the main cost drivers considered for HV and LV network investment. On the other, given that charges are to influence future behaviours, the investment costs in the determination of charges should be future network costs rather than historical costs.

This thesis presents the original contributions in the following aspects:

-A new charging model for HV distribution networks is developed. Future investment costs are identified by determining the required reinforcement in terms of the two cost drivers under a projected load growth. The investment costs are allocated among users according to their „contribution‟ into the costs to determine the charges. The underlying idea is grounded that costs should be allocated to those who cause them and by how much. The charging model provides locational and cost-reflective charges to users: the more costs they incur, the higher the charges are.

-A novel statistical model to quantify future investment costs is proposed for large-scale LV distribution networks. Due to the extensiveness of network configuration and limited approachable data, the triangular probability distribution is used to represent the distribution of utilisation levels of circuits and transformers in LV networks. The representation allows the assessment of the scale of network assets to be reinforced based on probabilities. The quantification of future investment costs provides a basis in developing a charging model for LV networks.

-Considering the increasing number of Microgeneration (MGs) connected at distribution networks, this thesis assesses the economic efficiency of LRIC in guiding future MG installation. A novel approach is introduced to quantify the investment deferral resulted from MGs for assisting the assessment.

-Since cost allocation theory always comes in terms of average cost and marginal cost, the debate on the choice between these two is on-going by researchers. In this thesis, preliminary analysis and discussion between these two cost reflective mechanisms is carried out for LV network charging models.

Date of Award	27 Jun 2012
Original language	English
Awarding Institution	University of Bath
Supervisor	Furong Li (Supervisor)