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

The authors of the present paper have developed a decentralized automation concept of LV-grids in recent work. When considering MV- or LV-grids, one of the most interesting aspects is a detailed overview of the current grid state almost in real time. Furthermore, grid state identification is the primary requirement for any continuative surveillance or control method. In contrast to state estimation on HV-level, a whole lot of different methods have been developed for calculation of grid states on MV- or LV-level within a sparse measurement environment related to cost minimization for measurement devices. However, most methods have not been developed for or tested on a decentralized automation system. The scope of the present paper is a comprehensive screening of available grid state identification methods and an adaption of these approaches with respect to the use within a decentralized automation system for MV-grids. In particular, five established methods for MV-grid state identification are applied to a representative 11-node MV-grid in order to compare the accuracy of the calculated grid state.