The “Sensors in the Grid 2.0” project delivered a comprehensive architecture for the secure acquisition, transmission and processing of voltage and broadband power‑line communication (BPL) spectrum data in low‑voltage distribution networks. Over 3,500 prototype BPL sensor modems were commissioned and connected to a scalable data‑processing system based on the NMS3 platform. The sensors performed real‑time, high‑resolution voltage measurements and captured full‑band BPL spectra, producing JSON‑formatted data streams that included signal‑to‑noise ratio (SNR) vectors and tone‑maps. A dedicated data‑preprocessing routine on the modems reduced bandwidth requirements by filtering and compressing raw measurements before transmission.

Laboratory validation at the Smart‑Grid Lab of the Bergische Universität Wuppertal and simulation studies confirmed the suitability of the overall system for network operation and asset management. The validation framework, illustrated in the project’s communication‑structure diagram, demonstrated that the FiN‑infrastructure could support automated network functions such as neighbor discovery and fault detection. A cost‑benefit analysis performed by PPC showed that, depending on rollout scenario, the BPL‑based FiN system could reduce network‑operation costs by 27 % to 38 % over an eight‑year horizon compared with conventional radio‑frequency solutions. Further analysis by BUW indicated that, for different network types, the innovative measurement approach could cut total costs by 60 % to 80 %.

The project also advanced data‑linking techniques that associate power‑grid topology with communication links. Using DBSCAN clustering on the SNR vectors, the DFKI team automatically identified relevant BPL links for grid‑state assessment and visualised temporal cluster evolution. This enabled a “labelling” of measurement data that is essential for training AI models. Early AI experiments applied clustering and classification to detect abnormal voltage patterns and predict equipment status, demonstrating the feasibility of machine‑learning‑based grid monitoring.

Regulatory analysis by BUW clarified that voltage measurements taken in cable distributors and local‑substation equipment are not personal data and therefore fall outside the scope of the GDPR. In contrast, data collected directly from end‑users are protected and must be anonymised or pseudonymised. The project also examined the eligibility of the FiN‑system for inclusion in network tariffs, concluding that the system’s net‑supporting and cost‑efficient characteristics satisfy the criteria for tariff‑eligible investments.

The collaboration was led by Power Plus Communications AG (PPC) and funded by the German Federal Ministry of Education and Research (BMBF). Key partners included the Bergische Universität Wuppertal (BUW), the German Research Center for Artificial Intelligence (DFKI), the Energy and Water Research Institute (EVL), Software AG, and the measurement‑equipment manufacturer WitiKee. The project ran from 2019 until its conclusion in November 2022, producing a final report that summarises the technical achievements, regulatory findings, and projected economic benefits of the FiN‑infrastructure for future distribution‑grid automation.