Models for generating place and time dependent urban energy demand profiles

In this paper, we present a new model for generating spatiotemporal power demand data for urban areas of the form P(x,y,t). The model is flexible and can be adjusted to different cases and local conditions. The dimensions of the model are not restricted, but a typical case would comprise an hour-by-hour simulation over a whole year with a spatial resolution from a few hundred meters up to several kilometers, depending on the area to be covered. These kinds of load profiles are useful when analyzing, e.g., smart grids, demand side management, and renewable energy in the urban context. The model was applied to two cities, Helsinki with detailed input data available, and Shanghai with access to rough data only. In both cases, the generated load patterns appeared logical in terms of empirical observations on how power demand behaves in space and time.