A statistical method for optimising wind power contribution in a diesel supplied network

Wind energy exploitation is likely to be economically viable in many cases where electricity is generated using diesel fuel. The economics of such cases are examined using statistical methods. The aim is to determine the optimum level of wind power capacity that should be installed in a diesel supplied network on the basis of known facts about the network and about the generation and local wind activity. A model for the load duration curve has to be assumed if not available. The theoretical basis of the method is first described and some details of the computer program to implement the method are given. A generalised study to assess the sensitivity of the solution to a number of parameters provides some interesting insights. Finally, the method is applied to all 45 Greek islands that possess autonomous diesel systems, and it is concluded that, with pessimistic assumptions, fuel savings of up to 22% and net savings of up to 9% are feasible.