Arc faults in direct current systems can be caused by animal bite, faulty connections (e.g., at terminal points) or hot
spots in photovoltaic modules.
These energy discharges must be detected very reliably and the system has to be safely switched off before fires can occur.
Erroneous tripping should be avoided as far as possible in order to maximize the availability of the DC micro power grid.

In addition, symbolic regression and classification will be combined with Compressed Sensing to create the basis
for a new technology for the detection of direct current arcs.

Within this project the focus is on photovoltaic systems and battery systems with various additional sources of
interference, such as EMC radiation on the lines, distributed DC/DC stages (module optimizers), module types and cable aging.