PredProy: Towards Efficient Ex-Ante Foundations of Weather Data for the Simulation of Smart Cities and Power Grids

Smart cities are integrated into the smart grid. Various adverse impacts on the power grid are heavily influenced by weather conditions, such as overproduction or underproduction of renewable energy, and the demand for electric heating and cooling in smart cities. To mitigate these effects operationally, they need to be anticipated. When the research objective is an ex-ante operational effect, we deduce that ex-ante weather predictions are a preferable choice for simulating power grids and informing agent decisions compared to traditionally used ex-post reanalysis data. However, when ex-ante data is used, the procurement is slow and the iteration speed often low. In this paper, we provide an overview of numerical weather forecasting and its output as input data for power and demand forecasting in smart grid and smart city environments. As we found no system appropriate for our use case, we propose an implementation based on our previous experience, fused with insights from industry experience and the MARS archive system. Our research could potentially prompt a transition towards using ex-ante weather forecasts as a foundation for more accurate exante power grid and smart city simulations and the management of adverse effects, such as congestion, outages, or blackouts.