Following the increased consciousness regarding global warming and other environmental issues, there is a growing urge in cities for a more rational energy supply and use in buildings. In parallel, the range of possible actions has broaden, from building refurbishment and construction options, through local energy production (solar panels, heat pumps) to the larger scale re-use of waste heat, green electricity production or trading. This evolution has increased the complexity of urban energy management, and more information regarding the energy efficiency of the energy demand and supply is required to evaluate possible improvements. Crude models fail to include all aspects of this complexity, while specialised simulation tools make global assessments and comparisons difficult.There is thus an opportunity for research to provide tools dealing with the increased complexity of urban energy flow to study the numerous options available to decrease the primary energy use of buildings and the related greenhouse gas emissions.Focusing on the scale of a few hundred buildings, we model in this paper the disaggregated energy flow, from the resources through networks and energy conversion system towards the energy use in buildings for distinct services, as an oriented graph. Combining the limited available data with existing simulation tools and energy consumption data, a new graph theory inspired simulation method is formulated. The resulting detailed picture of energy flow can give access to a large amount of information but, most importantly, will provide a unied tool to compare energy effciency scenarios regarding a broad range of technological aspects of energy demand and supply.