Detection of Misconfigurations in Power Distribution Grids using Deep Learning

The electrical energy system is undergoing major changes due to the necessity for more sustainable energy generation and the following increased integration of novel grid connected devices, such as inverters. To operate reliably in novel circumstances, as created by the decentralization of generation, power systems usually need grid supportive functions provided by these devices. These include control mechanisms such as reactive power dispatch used for voltage control. In this work, an approach for the detection of misconfigured (wrongly parameterized control curve, etc.) grid devices using operational data is proposed. By generating and analysing operational data of power distribution grids, a Deep Learning approach is applied to the detection problem given. An end to end framework is used to synthesize and process the data as well as to apply the machine learning techniques on it. The results offer insights into applicability and possible ways to improve the proposed solution and how it could be employed by grid operators.