Generating low-voltage grid proxies in order to estimate grid capacity for residential end-use technologies: The case of residential solar PV.

Due to data restrictions and power system complexity issues, it is difficult to estimate grid capacity for solar PV on regional or national scales. We here present a novel method for estimating low-voltage grid capacity for residential solar PV using publicly available data. High-resolution GIS data on demographics and dwelling dynamics is used to generate theoretical low-voltage grids. Simplified power system calculations are performed on the generated low-voltage grids to estimate residential solar PV capacity with a high temporal resolution. The method utilizes previous developments in reference network modelling and solar PV hosting capacity assessments. The method is demonstrated using datasets from Sweden, UK and Germany. Even though the method is designed to estimate residential solar PV grid capacity, the first block of the method can be utilized to estimate grid capacity or impacts from other residential end-use technologies, such as electric heating or electric vehicle charging. This method presents:•A method for estimating peak demand based on population density and dwelling type.•Generation of low-voltage grids based on peak demand.•Sizing of transformers and cables based on national low-voltage regulations and standards.

Dataset for generating synthetic residential low-voltage grids in Sweden, Germany and the UK.

Assessing grid capacity on national and local levels is important in order to formulate renewable energy targets, calculate integration costs of distributed generation (such as residential solar PV and electric vehicles). Currently, 70-96% of the residential solar PV installations in Germany and Italy are found in the low-voltage grid. Previous grid assessments have relied on grid data from individual low-voltage grids, making them limited to a few cases. This article presents synthetic low-voltage grid data from a reference network model. The reference network model generates synthetic low-voltage grids using publicly available data and national regulations and standards. In addition, the article presents data of residential solar photovoltaic hosting capacity in low-voltage grids. The datasets are high-resolution (1 × 1 km) and contains data on electricity peak demand, share of population living in apartments and important grid metrics such as transformer capacity, maximum feeder length and estimations of residential solar photovoltaic hosting capacity. Datasets on grid components are rare and the dataset can be used to assess grid impacts from other residential end-use technologies, and function as baseline for other reference network models.