Demand Scenario Generation for Distribution Grid Planning via Digital Twin Modeling

This paper presents a tool for generating mid-term demand scenarios to support planning studies in distribution networks. Developed within the framework of a digital twin model of the grid, the tool leverages a data-driven approach and real historical measurements from the system to forecast future consumption patterns. These load measurements are used to statistically identify peak consumption hours and to estimate mid-term demand trends through seasonal decomposition techniques. The resulting forecasts and statistical load behavior insights are then used to generate realistic consumption scenarios corresponding to potential grid stress conditions. The proposed methodology has been implemented and validated through a real case study involving a 3000 nodes rural distribution network. For validation, the generated scenarios and their power flow results are compared with those obtained using the actual demand.