Electrical networks and economies of load density

The objective of this thesis is to determine electrical network cost and cost structure depending on load density. Load density together with a defined customer mix directly determines the supply task, i.e., the number of connection points and their density in a geographical area as well as their load profiles. Load density indirectly determines the construction conditions since higher load density means a higher building efficiency which in turn leads to restrictions in land use and more strict structural requirements. The applied methodology adopts features from several engineering-economic approaches with the three basic steps: (1) Description of the supply task including relevant characteristics of the service area. (2) Network generation. Linear approximations of cost functions for lines and substations are used together with geometrical models of symmetrical and homogeneous load dispersion to determine optimal substation spacings. (3) Monetary assessment of the generated asset structure including operational and maintenance costs. The cost-effects of different external cost-drivers are incorporated into the applied cost functions. All network levels from low voltage connections up to EHV substations are included. Additionally, reliability indices are determined for the model network sample feeders with different mitigation strategies. Actual substation service areas are used as references to the calculated results. Additionally, based on the connection point data of these reference substation areas, another reference medium voltage network is created using an optimization algorithm developed at Helsinki University of Technology, Department of Electrical Engineering. Using the model network approach, cost is determined for networks in rural, suburban, urban and urban core zones. The cost structure and level varies from zone to zone depending on the allowed equipment, share of open space and customer mix. When moving towards the highest load densities in urban and urban core areas, the cost lowering caused by higher connection point density is attenuated due to the higher unit cost of equipment. The created analytic and geometric network model, even with massive simplifications, is proven suitable for estimating the network volume and cost for different zones and load densities. The applied methodology forms a qualified framework for further development of the model network approach