Considering the large amount of energy consumption in cities, two-thirds of the overall consumption, these latter have an important potential in terms of CO2 emissions reduction. Therefore, energy strategies are needed at a city level and consequently, adequate planning tools are required to support urban energy planners assess their decisions (e.g. which buildings are the best to refurbish). This paper presents an ontology based approach for urban energy planning support applied to building refurbishment and building-integrated solar PV planning. The adopted methodology is an iterative, incremental process, where each iteration leads to the integration of a new planning decision. The process starts by the identification of the actors whose interests are affected by the decision, then developing/ re-using computation models that provide answers for their questions. The different models are integrated using an ontology that models the parts of the city within the scope of the questions to be answered. The system is applied in a district (about 1200 buildings) in the city of Vienna. The adopted approach provides different actors with specific information to their points of view. Furthermore, the output is aggregated to a common level of abstraction, to be understood by all the actors. This approach is applicable to different cities, as the ontology also integrates extension and upgrade mechanisms that provide flexibility to cope with different data-availability contexts.