Building Applied Photovoltaics and Building Integrated Photovoltaic systems are promising opportunities to develop renewable technologies in urban environment. However, the photovoltaic field acts as a solar and aeraulic mask for the underlying building that impacts the building's energy needs for air conditioning and heating. In this work, the energy balance of a simplified building with integrated photovoltaic field is modelled using a Monte Carlo approach. This method is applied to evaluate the needs for cooling or heating of the building for several years. Two kinds of roof decks are studied: concrete roof with a single layer and green roof with two additional layers (substrate and vegetation layers). The results from the Monte Carlo simulation are discussed and exploited with a FeedForward Artificial Neural Network to interpolate the building consumption according to the meteorological parameters.