Real-time state identification for MV-grids on autonomous agent platforms

State forecasting in smart distribution grids: a modular approach using CARMA algorithm

“…Supported by a communication infrastructure based on IEC 60870-5-104 protocol, the unit is provided with continuous measurements in the monitored grid area and capable of sending commands to actuators (variable transformers, controllable loads or feeders). Just approximately 15 % of all nodes need to be measured for sophisticating results in the state estimation by network partitioning or sensitivity analysis [8]. Together with a parameter set, containing static information on the grid (branches, feeders and topology), the cyclical updated measurement builds up the input data base for the forecast algorithm.…”

“…Initial simulative tests of the implementation in combination with the smart grid monitoring unit have been applied on a grid model, equating a real medium voltage grid that consists of 104 nodes and 105 branches. 23 measurements were determined to be installed for the state estimation by network partitioning [8]. Only the apparent powers at measured points are considered as forecasted quantities, while there is no differentiation either it is a branch or feeder measurement.…”