Modelling network effects in a competitive electricity industry

To achieve a competitive, market based electricity industry, there have beenmany proposals for segregation of generation, transmission and distributionfunctions. However, most of these proposals do not treat demand side optionson an equal basis as supply side options. In addition, the presence of networkand its dynamics are usually not reflected adequately in the determination ofprices.Recently, the Nodal Auction Model (NAM) has been proposed. Its objective is tocreate a competitive electricity industry market model that incorporates theeffects of network and operation scheduling. For each trading period whichtypically lasts for half an hour, generators and consumers bid into an auctioncentre. Bids and offers for each trading period are resolved using an auctionmechanism with a network model embedded to maximise the benefit of trade.The main objective of this thesis is to develop alternative network models thatcould be embedded in the NAM auction process. Three network models aresuggested: the Exact model, the Transport model and the Piecewise LinearNetwork model. The Exact model results in a nonlinear problem and is used asa benchmark for the other two. The Transport model, which results in a linearprogramming problem, particularly suits a network with small variations in lineimpedances or a network which is radial in nature. The Piecewise LinearNetwork model, which also results in a linear programming problem, isfavourable for its speed and robustness as well as its ability to provide resultsclose to the Exact model results.The Exact model is developed further to include a commercial approach to thequality of voltage. Using this approach, generators and consumers include theirvoltage preference in their offers and bids submitted to the auction process. Asa result, the cost of maintaining voltage quality by the industry would bereflected in the nodal prices