Model Generation of Cyber-Physical Power Distribution Systems Using Land-Use and Open Data

The increasing interdependence between electrical power systems and infor- mation and communication technologies (ICT) is changing the operational landscape of modern energy systems. Accurate modeling of regional cyber- physical power systems (CPPS) is essential for assessing system performance and resilience, evaluating monitoring and control algorithms and designing communication-based services. However, the lack of available and repro- ducible interdependent system models remains a significant challenge. This paper addresses this gap by proposing a methodology for generating rep- resentative CPPS models. The methodology uses open geospatial datasets, such as land-use information and publicly available ICT and power infras- tructure data, to produce an ensemble of plausible CPPS models for the target region. It builds on existing approaches to generating logical power grid and ICT topologies, extending them with a genetic algorithm-based embedding method that spatially positions these networks based on regional land-use characteristics and observed interdependencies. The results discuss the char- acteristics of the power grid and the ICT access network across different land-use and scenario types. The proposed method reduces the objective function values from 34% of its theoretical maximum to 1.23% and lowers dispersion by 99.3% compared with the uniform baseline. The generated CPPS models show strong agreement with the open-data infrastructure and land-use characteristics, achieving average adherence values of 87.7% and 86.5%, respectively. The results indicate that the proposed approach can pro- duce representative CPPS model ensembles and capture their infrastructural characteristics with sufficient accuracy.