Distribution networks' observability: A novel approach and its experimental test

The work presents an approach devoted to improving the observability of distribution systems with a high share of Renewable Energy Sources. The proposed method, developed according to the prescriptions of Italian Resolution 646/2015/R/eel, but designed to be applied in any modern power system, provides the delivery to the Transmission System Operator of real-time estimates of the overall generation downstream an HV/MV substation. The estimation process is based on the real-time monitoring of a limited set of power plants, completed by the acquisition of weather nowcast and energy measures collected on users through the standard Automatic Meter Reading infrastructure. The approach, tested on two real Italian distribution networks, and benchmarked against an Artificial Neural Network based approach, showed a good accuracy, allowing the provision of useful information to the TSO through a simple and easy to implement architecture.